Knowledge Mapping of the Development Trend of Smart Fisheries in China: A Bibliometric Analysis

Abstract

In recent years, smart fisheries, as an emerging model for fishery development, have become a research hotspot in the fishery and aquaculture industries of many countries. Smart fisheries can be thought of as a system that combines techniques for raising, catching, or selling aquatic products to improve production and sustainable development. Smart fisheries are crucial to improving fishery and aquaculture management. In this study, a comprehensive analysis was conducted using bibliometric analysis, the results of which are presented through visual mapping and data charts. This study collected data from the China National Knowledge Infrastructure (CNKI) database and compared it with the WoS database. A total of 949 articles were retrieved on topics related to smart fisheries, including 579 articles from WoS and 370 articles from CNKI. The results present the visualization and analysis of annual publications, author collaboration maps, research collaboration institutions, keywords, etc. The development of smart fisheries in China is obviously different from that in foreign countries. China attaches great importance to technology and production, while foreign countries focus on environmental issues. Therefore, this study helps us to understand the current research status, research hotspots, and future development directions of smart fisheries, providing certain references for future management.

1. Introduction

The growth and competitiveness of early industries were based on the application of process control technology in production systems [1]. In the late 1970s and early 1980s, with the development of artificial intelligence technology, the world’s first agricultural production management expert system was born in the United States, mainly for the diagnosis of crop diseases and pests [2]. In the 1970s, with the rapid development of wireless electronic technology and information technology, the United States applied advanced science and technology to the traditional aquaculture model [3]. For example, Boston Seafood Company and Deep Blue Seafood Company in the United States use an automated online testing system to monitor water quality and fish growth. In the early 1960s, an automated monitoring system for water environmental factors was first applied to aquaculture in Gunma Prefecture in Japan [4]. In the late 1990s, agriculture became the pioneer of world productivity through intensification, mechanization, and automation. Similar trends and pathways currently apply to fisheries and aquaculture [5].

Since then, the first microcomputer-based control system has been used to control environmental conditions in marine fish spawning system costs [6,7,8]. The most important parameters that need to be monitored and controlled in aquaculture systems include temperature, dissolved oxygen, pH, ammonia, nitrates, salinity, and alkalinity, as they directly affect animal health, feed utilization, growth rate, and carrying capacity [9]. Aquaculture countries such as Norway, Canada, and the United States have developed and promoted a series of automatic feeding systems, which effectively improved the feeding intensity and feed utilization rate and reduced labor intensity and feed waste, thereby reducing aquaculture costs and residual bait residues [10]. Factory aquaculture systems in the United States in the 1960s and 1970s were used to rapidly develop factory fish farming, and were regarded by the U.S. government as “one of the ten best investment projects”.

At that time, there seemed to be no such thing as smart agriculture or smart fisheries in academia. Farmers are able to optimize crop yields and increase efficiency thanks to innovative agricultural methods that integrate technology into farming practices. This has led to the emergence of a new method called “smart agriculture” [11,12]. Intelligent aquaculture is understood as using the Internet of Things (IoT), big data, artificial intelligence, 5G, cloud computing, and robotics to independently control aquaculture facilities, equipment, and machinery through remote control or robots to complete all production and management operations [13]. Different from intelligent aquaculture, Vo mentioned that, in addition to technology and processes, smart aquaculture also emphasizes a sustainable perspective [14]. The smart fisheries literature over the past decade shows great progress in improving our understanding of the operation of artificial intelligence technologies in fish farming in the early years, while in the later years, there was a particular focus on optimizing the efficient use of resources in ecosystem management [15,16].

Sea surface temperature (SST) maps measured by satellites were widely used in the 1960s and 1970s by space agencies, meteorological centers, and fishery agencies in the United States, United Kingdom, Canada, Japan, Australia, France, Russia, and many other countries [17]. Over the past few centuries, the mechanization of fishing and other developments such as technological innovations in ship design, trade organization, and the development of transport facilities have allowed fishing activities to expand spatially [18]. At this time, many technological innovations such as ship construction and fishing gear materials and manufacturing, as well as the use of electronic equipment (e.g., radar and echo sounders) and efficient fishing gear (e.g., trawls) have brought fishing effort and fishing efficiency to unsustainable levels [19,20]. In the past, intelligent fisheries or modern fisheries seemed to show that advanced equipment made the fishing process more efficient. However, it often brings greater risks and greater environmental impacts [18]. As a result of this development, fishing-induced species extinctions or near-extinctions appear to be more frequent than previously thought (see the review by Roberts and Hawkins (1999)) [21], despite the application of fishery prediction models [22,23]. Smart fisheries at sea refer to a scientific field whose goal is to automatically count fish to monitor the health of fish communities and thereby optimize fish harvesting [24]. Known as Global Fishing Watch (GFW), it builds geospatial datasets, hosts an online mapping platform, provides new ways to track fishing activity, and is driving new developments in the management of fisheries [25].

The rapid increase in the amount of literature on smart fisheries over the past decade represents tremendous progress in the operation and understanding of smart technologies in fisheries and aquaculture [16,24,26,27]. Ebrahimi et al. [24] mentioned that the geographical distribution of authors demonstrates that the topic of artificial intelligence in fisheries systems is of interest to scholars worldwide. Mustafa et al. [15] reviewed smart fish farming systems and demonstrated the application of complex science and technology to aquatic production systems. Smart fisheries are crucial to improving the management of fisheries through innovative technologies and advanced data systems [28]. Advanced analytics will help bring direct benefits to fishermen by making fleet operations more efficient, for example, by allowing vessels to find target species faster and reduce fuel use and costs [29,30]. Therefore, in addition to the benefits of technological development to the industry, the use of smart fisheries management has also been widely developed.

Historically, the perspective of sustainable fisheries development has been that it plays an important role in socioeconomic progress and human well-being [31]. In addition to the issue of avoiding the overexploitation of fish stocks, where the concept of smart fisheries has been proposed, there are other issues related to fair trade, equality, and decision-making. In addition, it supports regulations and public policies that contribute to environmental health, economic prosperity, human well-being, equality, and justice [32]. Digital technologies are bringing significant operational benefits to the global food chain, improving efficiencies and productivity and reducing waste, contamination, and food fraud. Smart fisheries involve multiple social sectors such as academia, industry, fishermen communities, and non-governmental organizations, or the role of a Quadruple helix Hub (academia, industry, government, and society) [33]. The operation of smart fisheries creates knowledge, innovation, and technology, and at the same time, has the responsibility of raising awareness among fishing communities, governments, and consumers [34]. Smart fisheries should be considered a part of smart cities; therefore, efforts are also needed to ensure effective industrial management to advance social progress [35]. Regarding food safety, they can provide traceable food and ensure effective supply chain management [36,37]. Therefore, supervising and monitoring product quality through symmetric information traceability is very important in fresh food and fishery supply chain management. Aquatic food safety and traceability systems based on blockchains, the IoT, Wireless Sensor Networks (WSN), and Radio Frequency Identification (RFID) provide reliability from production to consumption [36]. For example, an Internet technology (IT)-based tuna traceability system was developed in Indonesia because the country is one of the major tuna exporters and has a complex supply chain network [38]. In addition, there are boxes containing fish and smart tags, so the instant status of seafood can be grasped when many tags pass through a fully automatic reader [39]. Yan et al. [40] designed and developed a traceability platform through the IoT, making it possible to trace aquatic food from breeding to processing, circulation, and sales.

Since China’s reform and opening up, the fishery and aquaculture industries have developed rapidly. In 2022, China’s total aquatic product output will be 68.6591 million tons, an increase of 2.62% from the previous year. Among them, aquaculture production was 55.6546 million tons, up 3.17% year-on-year, and fishing production was 13.0045 million tons, up 0.35% year-on-year [41]. Although the total volume of aquatic products in China has been rising significantly every year, there are many natural and man-made dilemmas in the aquaculture industry that need to be resolved, such as the scale of pollution of the marine environment and the serious eutrophication of the water quality caused by the shrinking of the aquacultural area. In addition, aquatic products convergence due to the increase in the number of types of aquatic diseases and the difficulty of preventing them [42]. Most of these problems exist in the aquaculture industry, such as eutrophication of aquaculture water, inadequate detection of dissolved oxygen in the water, irrational use of fish medicine, and inadequate technology promotion [43]. In addition, there are still a series of problems such as backward production methods, unreasonable industrial structures, backward deep processing technology of aquatic products, weak independent research and development capabilities of aquatic product processing equipment, a shortage of aquatic professional talents, and an uneven quality of employees [44].

In response to the above problems, in addition to some traditional methods, the new fishery model of “smart fishery” has developed rapidly in recent years. The construction, management, and operation of smart aquaculture systems have not only reduced fishery breeding costs and improved fishery economic benefits but have also promoted environmental protection and water quality. In the “Several Opinions on Accelerating the Green Development of Aquaculture” issued by 10 ministries including the Ministry of Agriculture and Rural Affairs of China, it was clearly stated that it is important to “promote smart aquaculture, guide the in-depth integration of modern information technologies such as the Internet of Things, big data, and Artificial Intelligence (AI) with aquaculture production, and carry out digital fisheries demonstration” [45]. In this regard, smart fisheries can be understood as the process of applying modern information technologies such as the IoT, big data, artificial intelligence, satellite remote sensing, and mobile Internet to deeply develop and utilize fishery information resources and comprehensively improve the comprehensive productivity of fisheries. Furthermore, improving the efficiency of fishery management is an important means and an effective way to promote the structural reform of fishery supply and accelerate the transformation and upgrading of the fishery [46].

In China, Zhang et al. mentioned the application of IoT technology in aquaculture water quality monitoring, aquatic product cold chain logistics, aquatic product quality and safety traceability, and marine fisheries in 2014 [47]. Based on the development of smart fisheries in Zhoushan, Ni [48] proposed six innovative models in 2015, including smart mariculture, smart breeding, smart ocean fishing, innovative smart services, and smart ecological fisheries. In 2017, Wang et al. proposed an innovative path of an intelligent “Internet + marine fisheries” model, which is the product of information productivity covering the entirety of the marine industry chain, through the optimization and restructuring of all aspects of fishery farming, to promote the transformation and upgrading of the fishery industry chain [49]. Yin et al. [50] studied the smart fishery development model in Hukou County, Jiangxi Province, which established a “1 + 4” smart fishery management model to promote the development of modern fisheries in 2018. In 2022, Wei [51] discussed the research on the application of machine vision technology in three aspects of fish farming: fish feeding behavior recognition and accurate feeding, fish body parameter recognition and measurement, and fish disease diagnosis and control.

Since fisheries have existed for a long time, the smart fisheries that emerged from these seem to have also shown complex academic development. Therefore, it is crucial to understand the development trends and current status of smart fisheries. Bibliometric analysis is an emerging field and is a popular and rigorous method for exploring and analyzing large amounts of scientific data [52]. Its popularity can be attributed to bibliometric software such as Gephi, Leximancer, VOS viewer, and Citespace, as well as scientific databases such as Scopus and Web of Science (WoS) [52]. Bibliometrics is a discipline that takes literature-related media as its research object and uses measurement methods such as mathematics and statistics to study the quantitative relationships and patterns of documents, as well as to explore the dynamic characteristics of documents [53]. Since it is a scientific computer-assisted review method, core studies or authors and their relationships can be identified by covering all publications related to a given topic or field [54].

Many Chinese scholars have published Chinese academic papers to understand the research status, development trends, and research frontiers of smart fisheries in China. However, Chinese articles are not easy for foreign scholars to understand. Therefore, comparison and discussion through the literature in different languages is important and meaningful. This study collected data from the CNKI database and then compared it with the English version of the WoS database, using Citespace as the visual analysis software to conduct knowledge graph analysis. Keyword co-occurrence diagrams, timeline distribution diagrams, emergence diagrams, etc., were mined from two different language databases for visual analysis, aiming to explore the current status and academic frontiers of smart fishery research in China. It is hoped that the results of this research can provide guidance for the future development of China’s smart fishery research field and provide a strong reference for the high-quality development of fisheries.

2. Materials and Research Methods

2.1. Data Sources

The databases used were WoS and CNKI. WoS was chosen because it covers more than 20,300 scientific journals, books, and conferences, as well as more than 71 million items of research material [55,56]. In addition, CNKI covers more than 90% of China’s knowledge information resources [57].

WoS is a multidisciplinary large-scale literature database developed by Clarivate Analytics, which has access to ISI’s three major citation databases (SCI, SSCI, and A&HCI) and two conference proceedings indexes (CPCI-S and CPCI-SSH), including the world’s high-impact journals and the conference proceedings of nearly 15,000 international conferences and symposiums in the fields of natural sciences, social sciences, humanities, and arts. The database also includes nearly 15,000 journals and 120,000 proceedings of international conferences.

The main function of CNKI is to provide search and reading services for all kinds of academic journals, dissertations, conference papers, newspapers, almanacs, encyclopedias, and other kinds of Chinese literature resources, etc. CNKI also provides some auxiliary tools, such as literature management tools, reference generation tools, and discovery of knowledge tools. The two databases, CNKI and WoS, were chosen as the source of data for this study, not only because the number of articles is large but also because the professionalism and authority of the articles are high, meaning that the results can be analyzed effectively and reliably.

In CNKI’s advanced search mode, with “fisheries” as the subject, plus “artificial intelligence” OR “internet of things” OR “big data” OR “satellite remote sensing” as the keywords, and “cloud computing” OR “internet of things” as the keywords, and “satellite remote sensing” as the keywords OR “intelligent fisheries” as the abstract, and finally “intelligent fisheries” as the subject OR “internet of things” as the abstract”, select the corresponding disciplines and carry out the matching search. Given that there are relatively few Chinese academic journals on smart fisheries in the China Knowledge Network (CNKI) database, the time scale of the research papers selected in this paper is set as 1994–2023, and the advanced search function in CNKI is utilized to eliminate non-academic papers and retain only “academic journals” for searching, with 370 papers finally being screened out. The obtained documents were exported using the Refworks format in China Knowledge Network and saved in the “download_.TXT” mode so that Citespace could be used for the visual analysis of the documents. The search time was up to 30 December 2023.

CNKI-specific level searches are listed below:

A.

Take “Fisheries” as the theme, add “Artificial Intelligence” as the keyword or “Internet of Things” as the keyword.

B.

Take “Fisheries” as the theme, add “Artificial Intelligence” as the keyword or “Big Data” as the keyword;

C.

Take “fisheries” as the theme, add “cloud computing” as the keyword OR “Internet of Things” as the keyword;

D.

Take “fisheries” as the theme, add “artificial intelligence” as the keyword OR “satellite remote sensing” as the keyword;

E.

Take “Fisheries” as the theme, add “Satellite Remote Sensing” as the keyword OR “Intelligent Fisheries” as the abstract;

F.

Take “Fisheries” as the theme, add “Intelligent Fisheries” as the title OR “Internet of Things” as the abstract;

In the WoS Core Collection database, with the subject set to =((((TS = (fishery)) AND TS = (Smart fishery)) OR TS = (Digital fishery)) OR AK = (Smart fishery)) OR AK = (Fishing forecast), a total of 579 journal articles between 1992–2023 were obtained from the WOS Core Collection database search. In data export mode, select plain text mode and choose full record and cited references for export, saving the export in the “download_.TXT” mode for subsequent visualization and analysis of the literature using Citespace. The search was open until 30 December 2023.

2.2. Research Methodology

Citespace is a multivariate, time-sharing, and dynamic knowledge mapping tool [58] that can intuitively display scientific knowledge, visualize abstract data, delve into the interior of the knowledge, and systematically analyze its developmental process and structural relationships to draw scientific conclusions. Researchers can have a more comprehensive understanding of their research field, keep abreast of the latest cutting-edge developments in the research field, quickly adjust the research direction, and predict the future direction of development. The literature analysis was carried out with the help of Citespace visualization software (version 6.2.4R), and the literature section used the duplicate weight reduction function in Citespace. For the software to recognize the imported data, it needs to be formatted after import.

2.3. Analysis Procedures

However, not all documents undergo a rigorous academic review process. For this reason, selected articles from both databases were manually read, filtered, and screened. Since the keyword “smart fishery” is not common, “Intelligent fishery” was also selected for reading and filtering. These manually selected articles with meaningful and correct content were then placed into folders for bibliometric analysis. In this sense, documents published in Chinese with titles, abstracts, and keywords were also included after this process.

Some scholars who use bibliometric methods often quantify and analyze parameters such as the time of publication of the literature, authors, citations, keywords, citation times, institutions of documents, etc., to systematically reflect the development trends in the research field and the frontiers of the discipline [24,56,59,60]. For example, for “Author” analysis, a metric related to the number of publications published over the years reflects interest in a specific topic. Indicators related to the number of publications each author has published over the years reflect their performance in the intellectual field [60]. The search for “Keywords” enhances the ability to search for cited references and can search across disciplines for all articles citing common references [60]. They can be used to compare their origins and are important parameters for extracting content and scientific concepts expressed in articles. According to Abafe et al. [61], using “author” and “keyword” can summarize the author and keyword usage frequency, which indicates the topic of interest and can infer the trend of topic evolution.

The number of publications by authors reflects trends in research topics and academic interest in smart fisheries. CiteSpace software (version 6.2.R4) was used to analyze and draw the knowledge graph of authors of smart fisheries publications in the CNKI and WoS databases. Each node in the graph represents a researcher, and the connecting lines between nodes represent the collaboration network of researchers. Then, the filtered documents are imported into CiteSpace software (version 6.2.R4) and the author (Author) node is screened. The selection criteria were the top 50 in each time zone, i.e., the authors with the highest number of citations. The time span was set to 1994–2023, the time slice was one year, k in the scale factor size index was set to 25, and Top50 was a single-year search threshold, using the default algorithm. The network clipping method was “Pathfinder + Pruning slicing network”. Finally, the statistics of the number of articles published by core authors and the co-occurrence map of core authors were obtained. According to Price’s law formula: m = 0.749$\sqrt{N_{max}}$ (N_max refers to the highest number of published articles). Therefore, higher values represent more productive and influential authors in the field.

The main forces in the field of smart fishery research can be understood by counting the number of publications by research institutions and universities. Select “Institution” as the node type, select 1 year as the time slice, and keep the others as default.

By analyzing keywords, research hotspots and frontiers in this field can be understood, which is of great significance for predicting future development trends. The node type was selected as “keyword”, the time span was 1994–2023, the time slice was selected as 1, the threshold N was selected as Top50, and the network clipping methods were selected as Pathfinder and Pruning sliced networks. The others were kept as default. Based on the network structure and clustering clarity, Citespace provides two indicators, the modular value (Q-value) and the average broad value (S-value), as the basis for judging the effectiveness of the mapping. The S-value is used to measure the homogeneity of the network, and the closer it is to 1, the higher the homogeneity of the network. If the value is above 0.5, the clustering result can be considered reasonable. When the value is 0.7, the clustering is efficient and convincing. Modularity denotes the network simulation blocking metrics; the larger the value, the better the clustering result of the network. If the Q value is between 0–1 and Q > 0.3, it means that the result of the delineated clusters is significant. The closer the value is to 1, the better the clustering.

In the analysis of institutions, the following are the relevant settings. “Institution” was selected as the node type, 1 year was selected as the time slice, and the time span was 1992–2023, with the others kept as default.

3. Results

3.1. Volume and Trends of Publications in WoS and CNKI

The number of academic publications over the years can show the development speed and status of the smart fishery discipline, reflecting the attention of scholars in this field. A total of 949 articles were retrieved on topics related to smart fisheries, including 579 articles from WoS and 370 articles from CNKI (Figure 1). As shown in Figure 1, the number of publications in the field of “smart fisheries” has shown an upward trend over time. Publication in Chinese journals seems to have developed significantly since 2012.

The overall trend of CNKI’s publications can be divided into two stages. Stage 1 was the starting phase (1994–2012), during which the CNKI database published 11 documents, with an average of 0.58 articles per year, which is a relatively small number of publications. The second stage was the rising stage (2013–2023), where the CNKI database published 359 documents, with an average annual volume of 32.64 articles per year.

The overall trend of WoS’s publication can be divided into three stages. Stage 1 was the start-up stage (1992–1999), where the WoS database published 10 papers with an average of 1.25 papers per year. Stage 2 was the exploration stage (2000–2012), during which the WoS database published 147 articles, with an average of 11.31 articles per year. Stage 3 was the rising Stage (2013–2023), in which the WoS database published 423 pieces of literature, with an average of 38.45 articles per year.

3.2. Analysis of Authors in the Research Field of “Smart Fishery” in CNKI

CiteSpace bibliometric visualization software was used to analyze and draw the knowledge graph of authors of smart fisheries publications in the CNKI database (Figure 2). Each node in the graph represents a researcher, and the connecting lines between nodes represent the collaboration network of researchers. The authors of this article imported the 370 selected documents into CiteSpace software to screen the author (Author) nodes. Finally, the statistics of the number of articles published by core authors (

Table 1. Top 25 authors with the most papers published on CNKI.

No.	Number of Publications	Year	Author	No.	Number of Publications	Year	Author
1	6	2013	Dao-Liang, Li	14	2	2016	Yi, Liu
2	4	2013	Liang-Liang, Ye	15	2	2017	Zhe, Yu
3	4	2014	Xing-Qiao, Liu	16	2	2021	Juan, Chen
4	4	2018	Feng, Liu	17	2	2016	Jian, Zou
5	3	2018	Yu-Qing, Liu	18	2	2023	Jin-Hua, Lin
6	3	2013	Shou-Qi, Cao	19	2	2021	Shou-Li, Xiong
7	2	2016	Jing-Hui, Fang	20	2	2013	Yong-Ming, Yuan
8	2	2017	Fan, Wu	21	2	2021	Xin-Miao, Pang
9	2	2006	Xue-Sen, Cui	22	2	2016	Sheng-Nan, Zhang
10	2	2019	Hui, Guan	23	2	2018	Jia-Jia, Li
11	2	2015	Yan-Zhong, Liu	24	2	2011	Qiang, Yao
12	2	2013	Jun, Xia	25	2	2016	Ming-Hua, Shang
13	2	2023	Mao-Chun, Wei

) and the co-occurrence chart of core authors were obtained (Figure 2).

As can be seen from Figure 2, the color of the node represents the year and time. The red node is 2023. At the same time, the size of the node indicates the number of papers published by the scholar. The larger the node, the more papers the scholar has published. The authors Dao-Liang Li, Liang-Liang Ye, Xing-Qiao Liu, and Feng Liu have published six, four, four, and four papers, respectively (Table 1). According to Price’s law formula, this study calculated that the threshold for the most productive authors in the field of smart fishery research is 1.83, meaning that authors who have published two or more publications are considered to be highly productive authors. It is shown here that higher values represent more productive and influential authors in the field. After statistical analysis, there are 25 authors and a total of 62 papers published, accounting for 16.76% of the total number of published papers. In addition, there is little cooperation and communication between authors, the research is relatively scattered, and the mutual citation rate is relatively low.

Among the above-mentioned authors, Li et al. applied some new technologies from the Internet of Things, artificial intelligence, fishery equipment, etc., to promote better development of the aquaculture industry [62,63,64,65,66]. Ye and Ye et al. established a set of maricultural Internet of Things system models, performed statistical analysis on the collected data, including a series of parameters such as pH, dissolved oxygen, ammonia nitrogen, and oxygen content, and designed water quality parameters suitable for aquaculture objects [67,68,69,70]. Liu and Zhang also started with water quality parameters and used IoT technology to design smart water quality environment monitoring systems [71]. The advantages of the system include being able to understand various parameters of breeding water quality anytime and anywhere, independently of time, geography, and other factors, while also being able to adjust parameters to reduce breeding risks [72,73,74].

3.3. Institutional Analysis of the “Smart Fisheries” Research Field in CNKI

The knowledge map of research institutions in the field of smart fisheries is shown in Figure 3. There are 13 institutions that have published more than three papers. These institutions form the core force in the field of smart fishery research (

Table 2. Top 13 organizations with the most publications on CNKI.

No.	Number of Publications	Year	Institution
1	13	2013	College of Engineering Science and Technology, Shanghai Ocean University
2	10	2013	College of Information and Electrical Engineering, China Agricultural University
3	8	2013	Xiamen Oceanic Vocational College
4	6	2017	Fishery Machinery and Instrument Research Institute, Chinese Academy of Fishery Sciences
5	4	2018	Yantai Institute of China Agricultural University
6	4	1997	Freshwater Fisheries Research Center, Chinese Academy of Fishery Sciences
7	4	2017	College of Marine Sciences Shanghai Ocean University
8	3	2014	College of Engineering Science and Technology, Shanghai Ocean University
9	3	2022	School of Information Science & Engineering, Dalian Ocean University
10	3	2015	Institute of Science and Technology Information, Shandong Academy of Agricultural Sciences
11	3	2017	Faculty of Mathematics and Computer Science, Guangdong Ocean University
12	3	2014	School Electrical and Information Engineering, Jiangsu University
13	3	2018	University of Chinese Academy of Sciences

). Among them, the School of Information Science & Engineering, Dalian Ocean University in China has published more articles recently. The College of Engineering Science and Technology, Shanghai Ocean University published the largest number of articles with 13 articles. Judging from the connections between nodes, there is less cooperation between institutions, and research institutions appear to be more dispersed and have similar research capabilities. In summary, academic cooperation between institutions seems to require strengthening.

3.4. Knowledge Mapping Analysis of Keywords in the “ Smart Fisheries” Research Area of CNKI

By analyzing the keywords, the number of network nodes is 297, and the number of connections between nodes is 519. This result can help us to understand the research hotspots and frontiers in this field and is of great significance for predicting future development trends. In Figure 4, we can see that Q = 0.6802 and S = 0.8785. Therefore, the mapping using Citespace is reasonable and efficiently convincing. In Figure 4, we can see that Q = 0.6802 and S = 0.8785. If the S value is above 0.7, the clustering can be considered efficient and convincing. A Q value > 0.3 indicates that the divided clustering results are significant and the clustering effect is good. Therefore, it is convincing that the correspondence using Citespace is reasonable and effective.

In Figure 4, each node corresponds to a keyword. The larger the node in the graph, the more references to that keyword. The brighter the color, the more popular the keyword has been in recent years. The radius of the node indicates the frequency of the keyword appearing in the imported data. The larger the node radius, the more times the keyword is cited and the wider its scope. Among them, the two keywords, “Aquaculture” and “Internet of Things”, are relatively prominent. In addition, the connecting lines between nodes indicate that keywords appear together in the same document. More connecting lines indicate a greater number of connections between each keyword, meaning that it is more popular in the field and can attract more attention.

Closely related keywords are clustered by the cluster map, and then each keyword is assigned a value, while the maximum value in the same cluster serves as their label. Based on the mapped keywords, a total of 35 clusters were generated, and the top 10 larger clusters were selected to analyze hot spots and trends, namely aquaculture (#0), Internet of Things (#1), smart fishing Tank (#2), dissolved oxygen (#3), fishery industry (#4), rural revitalization (#5), big data (#6), satellite remote sensing (#7), aquatic products (#8), and smart fishery (#9). Then, the frequency and year information of relevant representative keywords contained in each cluster are counted and tabulated (

Table 3. Clustering of keywords in the research field of smart fisheries in CNKI.

No.	Year	Clusters	Keywords	Frequency
#0	2016	Aquaculture	Aquaculture; Internet of Things technology; Intelligent Sensors; Water quality monitoring and control	47
#1	2017	Internet of Things	Internet of Things; Remote control; Intelligent management; Time synchronization; Intelligent services	40
#2	2018	Smart fishing tank	Smart Fishing Tank; Remote control; Aquaponics automatization; Single chip microcomputer	25
#3	2018	Dissolved oxygen	Dissolved oxygen; Intensity of illumination; Water quality testing; Detecting system; Illumination	20
#4	2016	Fishery industry	Fishery industry; Breeding; Processing; Intelligent; Cloud computing	20
#5	2020	Rural revitalization	Rural revitalization; Marine ranching; Recreational fishery; Government; Fishery industry; Industrial transformation	20
#6	2016	Big data	Big data; Cloud service; Data center; Data sharing; Scientific basis	20
#7	2010	Satellite remote sensing	Satellite Remote Sensing; Fishery forecast; Fishery resources; Artificial intelligence	19
#8	2016	Aquatic products	Aquatic Products; manage; Monitoring; Water Quality Information; Real-time	17
#9	2022	Smart fishery	Smart fishery; Information technology; Fishery industry; Fisheries development	14

). On this basis, a timeline map is drawn by arranging cluster labels sequentially on the timeline (Figure 5). The occurrence of a keyword in each row represents the keyword that appears more frequently in that cluster and the first year that the keyword appears in the selected database. The lines between keywords indicate that they appear in the same literature.

The emergence of new words may represent new research directions in this field. Emergent words are keywords that appear frequently in articles published within a short period of time. The frequency changes of keywords within the selected period reflect research trends and frontiers. Based on keyword co-occurrence mapping, the 20 most prominent keywords are selected and plotted (Figure 6). Sorted in chronological order, the red area indicates that the keyword is in a hot research period. A red horizontal line mark is formed from the beginning to the end of the occurrence, indicating that the keyword is in the important area of concern. The longer the keyword hot spots appear, the longer the research front has been focused on.

Combining the above with the visual results, it is shown that smart fishery research in China can be divided into two stages of development.

In the first stage (1994–2012), the research was relatively simple. Research in this period mainly focused on the application of satellite remote sensing in fisheries and the use of network technology to understand the latest developments in fisheries. The clustered keywords are satellite remote sensing (#7), aquaculture (#0), and Internet of Things (#1). The keywords are “water quality monitoring”, “IoT technology”, “remote control”, “artificial intelligence”, etc.

Fan et al. summarized the research, application, and development of satellite remote sensing technology in marine fishery environment analysis and fishing production, especially in marine fishery conditions and forecast services [75]. Wang and Ge focused on the use of Web servers, Windows NT operating systems, and IIS Internet management systems. This technology can help fishermen quickly query fishery statistical information and understand fishery statistical information in a timely manner [76]. Zhi and Jun developed an aquaculture IoT smart control and management system that can monitor dissolved oxygen, pH, water temperature, etc. These achievements include the timely adjustment of water quality in crab ponds, as well as predicting the incidence of various diseases and taking protective measures, giving breeding objects a better growth environment and helping breeding users create more income [77].

In the second stage (2013–2023), IoT technology is increasingly used in fisheries and aquaculture, developing in the direction of intelligence and automation. The clustered keywords are aquaculture (#0), IoT (#1), Intelligent Fish Tank (#2), dissolved oxygen (#3), Fishery industry (#4), Rural revitalization (#5), Big Data (#6), Aquatic products (#8), and Smart fishery (#9).

In 2013, Lin [78] built an online monitoring system for marine aquaculture that can provide water quality parameters and make appropriate adjustments according to the situation. This system has the advantages of immediacy, intelligence, and humanity. Xu [79] proposed a smart aquaculture environmental monitoring system designed using Internet of Things technology and 3G technology. It can monitor water quality in real-time and realize the mechanization, automation, intelligence, and energy-saving management of aquaculture. In 2016, Tong et al. [80] inspected the first smart fishery demonstration base in Lichuan County, Fuzhou City, Jiangxi Province. The report showed that the base has many functions, such as breeding information management, online water quality monitoring, automatic control, video display, and multi-platform terminal control [80].

In 2019, a document released by 10 ministries and commissions of the Ministry of Agriculture and Rural Affairs (MARD) proposed “recommending smart aquaculture, guiding the in-depth integration of modern information technologies such as the IoT, Big Data, and AI with aquaculture production, and carrying out demonstrations of digital fisheries” [81]. Regarding this research issue, scholars have put forward some insights into existing problems in the development process of smart fisheries, moving towards the development of multiple farming models and the transformation of the fishery industry.

In 2020, Wang and Ying [82] put forward some suggestions, including the transformation and upgrading of the fishery industry, the introduction of new aquaculture technologies, and the construction of Internet + Aquaculture Intelligent Fisheries in response to some of the problems facing the development of the fishery industry. Shen [83] introduced a method to organically combine marine fisheries with the Internet to analyze and explore the innovation path of the smart ocean in 2021. Ya and Ma discussed the current situation and problems in the development of smart fisheries, and put forward some suggestions, such as increasing policy support, increasing technological research and development, data sharing, and other suggestions, aiming at providing strong support for the sustainable development of the fisheries industry in 2023 [84]. In 2023, Wang et al. believed that there were problems such as unclear development models, low application levels, and lack of industrial standards in the development process of smart fisheries; therefore, they put forward suggestions for improvement based on case studies [85].

Through keyword co-occurrence mapping, the 21 most prominent keywords are selected for mapping, sorted by time. The red area indicates that the keyword is in the research hotspot period. In the first stage, satellite remote sensing is the main keyword, while water quality monitoring, intelligent monitoring, and monitoring system are the main keywords. In the second stage, aquaponics, smart fish tanks, and other aquaculture modes are the research hotspots.

4. Discussion

4.1. Comparison and Analysis of Authors by WoS

Judging from the number of articles published by scholars analyzed in the WoS database, there are a large number of influential authors in this field, as shown in

Table 4. Clustering of keywords for smart fisheries research areas in WoS.

No.	Number of Publications	Year	Author
1	6	2021	Uttam Kuma, Sarkar
2	6	2020	Robert, Arlinghaus
3	4	2021	Valerio, Sbragaglia
4	4	2015	Eric, Delory
5	4	2007	Patrice, Brehmer
6	4	2012	Md Yeamin, Hossain
7	4	2012	Jun, Ohtomi
8	3	2012	Elgorban M., Abdallah
9	3	2009	Yinlin, Chen

. After statistical analysis, scholars with more than two articles had 166 articles, accounting for 28.6% of the total number of articles. In addition, the knowledge map of authors in the field of smart fisheries is shown in Figure 7. The infrequent appearance of Chinese names suggests that Chinese scholars have not published many articles in English-language journals on the subject. Only Dr. U. Chen appears in Figure 7. Obviously, for Chinese scholars, it is insufficient to show the development of this research field, so there is a need to strengthen academic cooperation and exchanges and publish articles in international journals.

Among scholars, Dr. U.K. Sarkar serves as a Director of ICAR—National Bureau of Fish Genetic Resources, Lucknow, and is a leading scientist with significant contributions to fish conservation biology, endangered species breeding, aquatic habitat modeling, climate-adapted inland fisheries, vulnerability assessments, cage culture technology, reservoir and wetland fisheries, etc. Through climate analysis, he and his team identified six climate-smart fisheries strategies in West Bengal’s floodplain wetlands, namely Temporary pre-summer enclosures, Submerged branch pile (Kata) refuges, Autumn stocking, Torch light fishing, Deep pool (Komor) refuges, and Floating aquatic macrophyte refuge fisheries (Pana chapa). These climate-smart strategies can be used to adopt sustainable climate-smart fisheries management in floodplain wetlands [86].

As part of a team of scholars, Dr. R. Arlinghaus described the importance of big data information for recreational fisheries. In the future, big data information is likely to become reference information for recreational fishery managers to make decisions [87,88,89]. The team of Dr. V. Sbragaglia [90,91,92] outlined the importance of technological advances and datatization in advancing recreational fisheries in the face of climate change, continued biodiversity decline, and changing social values. Dr. E. Delory specializes in instrumentation and signal processing for biomedical and oceanographic observation systems (Paris XII). He and his team have integrated fixed and mobile observation platform systems and are committed to the sustainable development of marine science [93].

4.2. Comparison and Analysis of Institutional Data Using WoS

In order to compare the differences in the number of publications between Chinese institutions and non-Chinese institutions, the database collected using the WoS database was also used. A total of 13 institutions produced more than eight publications. Five institutions ranked among the top five, namely the Indian Council of Agricultural Research (ICAR), National Oceanic Atmospheric Admin (NOAA)—USA, ICAR—Central Inland Fisheries Research Institute, Centre National de la Recherche Scientifique (CNRS), and Ifremer (

Table 5. Top 13 institutions with the most publications in WoS.

No.	Number of Publications	Year	Institution
1	22	2013	Indian Council of Agricultural Research (ICAR)
2	19	2000	National Oceanic Atmospheric Admin (NOAA)—USA
3	12	2013	ICAR—Central Inland Fisheries Research Institute
4	12	2004	Centre National de la Recherche Scientifique (CNRS)
5	10	2003	Ifremer
6	9	2015	Consejo Superior de Investigaciones Cientificas (CSIC)
7	9	2005	Chinese Academy of Sciences
8	8	2003	University of California System
9	8	2014	Consiglio Nazionale delle Ricerche (CNR)
10	8	2001	Commonwealth Scientific & Industrial Research Organisation (CSIRO)
11	8	2013	Chinese Academy of Fishery Sciences
12	8	2015	CSIC—Instituto de Ciencias del Mar (ICM)
13	8	2015	CSIC—Centro Mediterraneo de Investigaciones Marinas y Ambientales (CMIMA)

). Two Chinese institutions ranked seventh and 11th, respectively, namely the Chinese Academy of Sciences and the Chinese Academy of Fishery Sciences (Table 5).

The knowledge map of research institutions in the field of smart fisheries is shown in Figure 8. These institutions constitute the core force in the field of smart fishery research and have made the greatest contribution to smart fishery research. Some institutions will have more frequent exchanges and cooperation with other institutions, such as National Oceanic Atmospheric Admin (NOAA)—USA, Consiglio Nazionale delle Ricerche (CNR), Consejo Superior de Investigaciones Cientificas (CSIC), Centre National de la Recherche Scientifique (CNRS), etc. Despite this, there is still relatively little cooperation and exchange between institutions, especially between the Chinese Academy of Sciences and the Chinese Academy of Fishery Sciences, and with foreign institutions. If institutions can frequently conduct academic cooperation, it will be more conducive to the development of the research field.

4.3. Comparison and Analysis of Keywords Using WoS

By analyzing the keywords in the WoS database, we can understand the research hotspots and frontiers in this field. This analysis is also performed to compare the results of CNKN and understand the differences. In Figure 9, when Q = 0.7586 and S = 0.9127, the number of network nodes is 594 and the number of connections between nodes is 1351. These nodes have become a hot topic in the research field in recent years. Among them, five keywords: “fishery”, “management”, “climate change”, “abundance”, and “fish” are more prominent. Comparing the two keywords “aquaculture” and “Internet of Things” in the CNKI database shows that the hot topics in the research field are different.

According to the mapped keywords, after running Citespace software, a total of 45 clusters were generated. The top 10 larger clusters were selected to analyze research hot spots, trends, and directions. These are Citizen science (#0), Climate change (#1), Deep learning (#2), Biomass (#3), Fisheries management (#4), Fisheries acoustics (#5), Blue economy (#6), Alaska (#7), Condition factor (#8), and Fishing effort (#9). Then, the frequency and year information of relevant representative keywords contained in each cluster are counted and tabulated (

Table 6. Clustering of keywords for smart fisheries research areas in WoS.

No.	Year	Clusters	Keywords	Frequency
#0	2014	Citizen science	Citizen science; Recreational fisheries; Deep learning; Community-based monitoring; Community mapping	56
#1	2016	Climate change	Climate change; Ensemble forecasting; Temporal changes; Species distribution model	49
#2	2017	Deep learning	Deep learning; Artificial intelligence; Smart fishery; Convolutional neural network; Artificial fish swarm algorithm	39
#3	2008	Biomass	Biomass; Recruitment; Anadromous alewives; Lake; Aquatic vegetation	32
#4	2014	Fisheries management	Fisheries management; Optical sensors; Underwater sound; Multifunctional ocean sensors; Standards development	32
#5	2007	Fisheries acoustics	Fisheries acoustics; Fishery; Fish species identification; Digital data	32
#6	2007	Blue economy	Blue economy; Active packaging; Circular economy Fish side stream; Smart sensors	31
#7	2010	Alaska	Video surveys; Alaska; Coral reef fishery; Sea expeditions	29
#8	2013	Condition factor	Condition factor; Length-weight relationship; Multimoment statistical analysis; Statistical forecasting	28
#9	2010	Fishing effort	Harvest estimation; Fishing effort; Recreational fishing; Seasonal forecast; Habitat; Nearshore fishing	22

). On this basis, a timeline map is drawn by arranging cluster labels sequentially on the timeline (Figure 10).

A comparison of the top 10 larger clusters in the CNKI database (Table 3) shows that the hot topics in the research field are different. The results in China show that scholars focus on more technological developments, such as aquaculture, the Internet of Things, smart fishing tanks, dissolved oxygen, the fishery industry, rural revitalization, big data, satellite remote sensing, aquatic products, smart fishery, etc. Foreign scholars, on the other hand, pay more attention to environmental issues. This keyword analysis shows a big difference. From the top 10 larger clusters in the WoS database (Table 6), it can be seen that scholars are concerned about Citizen science (Frequency = 56), which ranks first, and Climate change (Frequency = 49), which ranks second.

The field of citizen science is notable, with thousands of volunteers actively involved in ocean research activities. These volunteers (also known as “citizen scientists”) bring their experience in participating in research, including extensive ocean surveys, and provide meaningful data [94]. Data can be used in the field of citizen science to provide effective information for bird management decisions [95]. Citizen science can also involve input and participation from a variety of stakeholders to identify resource issues, develop strategies, and increase trust among stakeholders [96]. Thiel et al. [94] presented a review focusing on marine citizen science data covering topics including micro- and macroalgae, seagrasses, coral reefs, and most major animal taxa, including fisheries management. In order for resource management agencies to meet the growing need for reliable data on fish stocks and their habitats, Bonney et al. [97] mentioned that in the South Atlantic region of the United States, the South Atlantic Fishery Management Council has begun to establish a citizen science program. The purpose of this project is to improve the quantity and quality of data for fishery management decisions and to build trust and promote mutual understanding among those involved to achieve long-term sustainable development. In this case, a citizen science program, Send Us Your Skeletons (SUYS), was developed to effectively monitor population recovery. SUYS asks fishermen to voluntarily donate fish skeletons of their caught species so that scientists can extract biological data, generate age structures, and conduct population assessment analyses [98].

As mentioned previously, the emergence of new words may represent new research directions in this field stemming from keywords that frequently appear in articles published within a short period of time. Research trends and frontiers can be reflected in keyword frequency changes. Based on the keyword co-occurrence mapping, the 20 most prominent keywords were selected and plotted (Figure 11).

Based on the results above, the research area of “smart fisheries” in WoS can be divided into two stages.

In the first stage (1994–2010), The keywords clustered in this phase are Fisheries acoustics (#5), Blue economy (#6), Alaska (#7), and Fishing effort (#9). As shown in Table 6, this stage mainly focuses on the use of technology in fishery farming. During this time period in China (1994–2012), research mainly focused on the application of satellite remote sensing in fisheries and the use of network technology to understand the latest developments in fisheries. For example, the clustering keywords are satellite remote sensing (#7), aquaculture (#0), and Internet of Things (#1). Keywords include “water quality monitoring”, “Internet of Things technology”, “remote control”, “artificial intelligence”, etc.

Among scholars, Yoshitomim and Embutsu [99] developed an automatic feeder by processing images of fish feeding behavior to increase the number of feeding times, reduce feed wastage, and diminish aquaculture costs and pollution of aquaculture water quality in 2002. Tango investigated the use of ozone in recirculating aquaculture systems to control pathogens and improve process water quality while minimizing bromate formation in culture water in 2003 [100]. In 2006, Srithongouthai et al. [101] developed a microbubble generation system (MBGS) to control dissolved oxygen levels suitable for fish farming, and the system was tested for its capability to operate underwater. Compared to traditional bubbling units, MBGS can efficiently add dissolved oxygen to water and reduce the cost and energy aspects of water aeration. Alver et al. [102] proposed a model-based feed-forward and feedback control system for maintaining feed densities at a desired level or following a certain trajectory, a model that could reduce manual labor and improve the stability of rearing conditions in commercial hatcheries in 2008. Haron et al. [103] proposed architecture for the implementation of a water quality monitoring system for aquaculture, which can remotely control water quality via the Global System for Mobile Communications (GSM). At the same time, the system detects the deterioration of water quality in aquaculture and sends alert messages via Short Message Service (SMS) to remind us to take appropriate measures.

In the second stage (2011–2023), the keyword clusters in this stage include Citizen science (#0), Climate change (#1), Deep learning (#2), Digital terrain model (#3), Fisheries management (#4), and Condition factor (#8), as shown in Table 6. During this time period in China (2013–2023), IoT technology was increasingly used in fisheries and aquaculture, developing in the direction of intelligence and automation. The clustered keywords are aquaculture (#0), IoT (#1), Intelligent Fish Tank (#2), dissolved oxygen (#3), Fishery industry (#4), Rural revitalization (#5), Big Data (#6), Aquatic products (#8), and Smart fishery (#9).

This phase of the study focuses mainly on research into the management of fishery farming models. In 2015, Luo et al. [104] discussed accurate and automated algorithms based on machine learning techniques combined with statistical methods for recognizing and counting fish in video clips of fishing operations. Clough used a recirculating aquaculture system (RAS) at VicInAqua to develop a pilot Nile tilapia hatchery in Kisumu, Kenya. This practice has helped the region to improve food security and reduce dependence on imported fish and wild stocks in 2020 [105]. Manoharan et al. [106] discussed an IoT-based system for integrating Improved Decision Machine Learning Algorithms (IDML), which is easy to connect and low-cost in 2020. In 2021, Ebrahimi et al. [24] called on us to enter into deeper aspects of future research on AI for sustainable fisheries, from the needs of fishermen, government policies, etc., in order to better promote the value and potential of smart fishing fisheries in different regions. Chukkapalli et al. [107] introduced a secure smart fishery ecosystem to protect internet-connected sensors from potential cyber-attacks and proposed various AI applications to help farmers manage their fisheries effectively in 2021. In 2022, Ristolainen et al. [108] discussed the use of a cost-effective flow meter, hydro mast, as a monitoring device for the flow of water in agricultural nets, which can inexpensively measure tidal currents passing through the aquaculture nets without disturbing the objects in the aquaculture nets and provide a comprehensive view of water surface conditions.

4.4. Smart Fisheries Promote Industry Performance

4.4.1. Smart Aquaculture Performance

The main difficulty faced by the aquaculture industry in production is the lack of integration of systems capable of monitoring environmental variables and animal physiological characteristics [109]. This situation can be seen in most aquaculture farms where human operators are still responsible for long periods of manual data collection. Management and decision-making appear to depend on the experience of the operator and may be negatively affected by human subjectivity [110]. Industry 4.0 can be defined as the most modern and automated industrialization and is described as a set of technologies that bridge the digital world through cyber–physical systems [111,112].

Applying Industry 4.0 concepts in aquaculture, using interconnected sensors and other automated equipment to instantly collect and store data, can improve the efficiency of farming operations and enable faster decision-making [112]. Through a review of the literature, Biazi clearly stated that the advancement of smart systems has improved the performance of the aquaculture industry [112]. Specific technologies are implemented, such as photography systems for animal and environmental monitoring, sensing networks and the Internet of Things for collecting, transmitting, and storing data, and automated equipment to support certain activities and monitoring tasks, as well as supporting data interpretation and decision-making. Tools and modern sensing strategies are used for hardcoding. A smart system has adaptive behavior that is able to handle new situations and environments and allows it to use the acquired knowledge to make decisions based on fuzzy logic, neural networks, or other techniques [112]. Operational processes in the aquaculture industry transition from experience-based tools to knowledge-based tools [113]. At the same time, manual operations were converted into automation, and the introduction of this new knowledge produced huge increases in production efficiency [14,114,115,116].

4.4.2. Smart Fishing Performance

The industrialization of fisheries in the 19th and 20th centuries brought technological innovations such as steam-powered ships, onboard refrigeration and catch freezing, synthetic netting materials, and information technology to aid communication, navigation, fish location and transportation, and monitoring of fishing gear performance [117]. As a result, these innovations led to the rise of larger vessels and fishing gear, making it possible to harvest fish stocks at previously inaccessible ocean locations and depths, greatly increasing productivity levels [118]. In the fisheries sector, “innovation” refers to the development and adoption of new ideas, technologies, practices, and methods to improve the sustainability, efficiency, and overall performance of fisheries [118].

Clearly, this technological advancement has facilitated the growth and development of fisheries; however, it has unfortunately also led to overfishing and ecological impacts in fisheries. Since the introduction of the United Nations Convention on the Law of the Sea (UNCLOS) in 1982, countries have committed to better managing and conserving fisheries, ecosystem-based approaches such as marine protected areas (MPAs), and reducing illegal, unreported, and unregulated (IUU) fishing to sustainably develop fish resources. The application of new technologies since then has enabled governments to collect more data on fish populations, better monitor, enforce, and evaluate fishery activities, and improve the effectiveness of policies to sustainably manage fisheries [119].

Therefore, recent fishery innovations have turned towards promoting sustainable practices, which have been formed to greatly reduce ecological and environmental impacts [113,120,121]. For example, Ingólfsson et al. [122] mentioned that in order to control the catch, a catch limit system was developed and tested. They investigated technical solutions to avoid excessive catches, and the associated loss of catch quality, in the blue whiting (Micromesistius poutassou) Northeast Atlantic pelagic trawl fishery. Although there are some studies dedicated to improving the selectivity problem [123,124,125], precise selectivity efficiency is still difficult to achieve. Nonetheless, Hilborn et al. [121] argued that through technological innovation and careful management, all types of fishing gear can be fished sustainably.

In addition, these technologies include increased computing power in handheld devices; the proliferation of user-friendly Global Positioning System (GPS) and Global Navigation Satellite System (GNSS) applications; increased ability to store, share, and analyze “big data”; drones and Diversity and greater durability of low-maintenance radar stations; accessibility and accuracy of satellite imagery; continued improvements in on-board digital cameras and video recorders; expansion of Automatic Identification Systems (AISs) and Vessel Monitoring Systems (VMSs); and offshore Internet use [119]. For example, imaging systems on autonomous underwater vehicles (AUVs) that perform visual inspections of complex habitats can provide a better understanding of fish populations. Additionally, AIS and satellite systems already monitor the way vessels move beyond the 12-nautical-mile territorial waters [119,126]. Governments, enterprises, social groups and organizations, and individuals are increasingly turning to these new tools.

5. Conclusions

Based on bibliometric analysis using Citespace software, this article retrieved 370 and 579 academic journal articles from two core databases, CNKI and WoS, respectively. Publication volume, authors, research institutions, keyword clustering, and keyword timeline mapping were systematically compared. Judging from the overall upward trend in the number of publications on smart fisheries, it shows that scholars are increasingly interested in research on smart fisheries, whether in China or other countries. From the point of view of authors and institutional cooperation, the cooperation of research institutions in Europe and the United States is still relatively close compared to China’s, and most of the results of the research are based on research units and colleges and universities. Collaborative research among Chinese scholars is poorly connected and relatively independent, and institutions are dominated by universities and research institutes, which are relatively scattered. In terms of keyword clustering and timeline mapping, the trends and frontiers of technological development can be seen. The results of the institutions show that European and American countries have long applied high-tech means to fisheries and aquaculture. China’s smart fishery research started late, which also shows that the level of science and technology is relatively low. In promoting the high-quality development of China’s modern fisheries, suggestions include taking advantage of local resources, continuing research on all aspects of smart fisheries, introducing and innovating technologies, and providing good policy conditions.