Conceptual Design of the Intelligent Electronic Monitoring and Reporting Model for Combating Global Illegal, Unreported, and Unregulated Fishing

Abstract

Preventing illegal, unreported, and unregulated (IUU) fishing—which depletes fishery resources—is a critical task in international fisheries governance. Many countries operate vessel monitoring systems (VMS) and electronic reporting systems (ERS) to track their fishing vessels, while regional fisheries management organizations (RFMOs) are actively considering the adoption of electronic monitoring systems (EMS). Although ERS and EMS share the same operational goals, differences in their concepts and functions lead to technical and institutional limitations when implemented separately. This study presents a conceptual design of an intelligent electronic monitoring and reporting (I-EMR) system model, which integrates the strengths of both systems to address these limitations and provides a framework for efficient operation. The necessity for the prompt and proactive adoption of such systems is reinforced by recent analyses of global IUU fishing trends, which indicate that IUU activities are not decreasing despite existing monitoring efforts. While empirical validation is beyond the scope of this study, the conceptual framework aims to support transparent management of fishery resources, facilitate real-time monitoring of fishing activities, and serve as a foundation for future pilot testing and operational deployment.

1. Introduction

Securing the sustainability of fishery resources by preventing illegal, unreported, and unregulated (IUU) fishing is one of the major challenges facing the international community today [1]. According to one study, more than eight million to 14 million tons of unreported fish catches are traded illicitly every year, costing the legitimate market between USD 9 billion and USD 17 billion in trade each year [2]. Some scientists suggest that if current rates of depletion persist, most large predatory fish stocks will have collapsed by 2048 [3]. However, IUU fishing has not been eradicated. Since IUU vessels began to be officially listed by RFMOs in 2002, an average of 9.52 vessels have been registered annually, with 16 vessels listed in 2024 [4]. As the social and environmental consequences of IUU fishing become more apparent, the global community is making multifaceted efforts to curb indiscriminate overfishing and conserve fishery resources, with the goal of sustaining marine ecosystems [5]. The international community has incorporated IUU fishing regulations into various international instruments aimed at its eradication, including the 1982 United Nations Convention on the Law of the Sea (UNCLOS), the 1993 FAO Compliance Agreement, and the 1995 United Nations Fish Stock Agreement (UNFSA) [6]. The primary direct monitoring measures include the vessel monitoring system (VMS) and the deployment of observers on board fishing vessels. Placing observers on vessels to monitor fishing activities has inherent limitations [7]. To address these limitations, since the early 21st century, electronic monitoring (EM) has emerged as a cost-effective complement to traditional fisheries monitoring [8]. EM can enhance the traceability of distant water fishing, analyze monitoring footage to verify catch and catch per unit effort (CPUE), and assess compliance with regulations [9]. In particular, active discussions on EM standards are being led by the Western and Central Pacific Fisheries Commission (WCPFC) and the Inter-American Tropical Tuna Commission (IATTC) [10]. Digital traceability and transparency in global fisheries highlight the need for system architectures that integrate multiple data streams—such as EM, ER and VMS—into a unified platform [11]. EM also holds promise as a powerful future tool when integrated with existing data collection programs [12]. Moreover, while EM, ER, and data management systems can enable cost-efficient fisheries management, their successful implementation requires strategic program design, the use of artificial intelligence, and system integration [13]. Recent studies reveal that EM systems record up to three times more retained catch and detect unreported discards compared to vessel logbooks, highlighting the limits of self-reported data and the need for integrated monitoring to improve accuracy and transparency in fisheries management [14]. Accordingly, this study aims to examine the requirements for introducing electronic monitoring systems (EMS) as called for by the international community and to propose the I-EMR system—combining EM and electronic reporting systems (ERS) functionalities—as a systematic and advanced monitoring framework.

2. Literature Review and Methodology

2.1. Literature Review

As shown in

Table 1. Previous studies on the effectiveness of EM and ER.

Article Title	Method of Study	Key Conclusions and Implications
“EM in fisheries: Lessons from global experiences and future opportunities” [8]	Analysis of global EM application cases	EM, as a monitoring tool, demonstrates several strengths that outweigh its weaknesses. It also holds promise as a future tool when integrated with existing data collection programs.
“EM for improved accountability in Western Pacific tuna longline fisheries” [14]	Evaluation of EM’s effectiveness in Pacific longline fisheries	EM data recorded greater species diversity than logbook entries. Its further expansion could enhance the management of both target and bycatch species.
“Digital transformation of Egyptian marine fisheries: A pathway to sustainable fisheries management” [15]	Assessment of MCS technologies in Egypt’s Mediterranean fisheries	The successful adoption of advanced monitoring systems requires a phased approach, beginning with targeted pilot programs in priority fisheries and followed by incremental scaling supported by technical capacity building.
“Improving ER Rates in the U.S. Recreational Fishery for Atlantic Bluefin Tuna” [16]	Survey on ER awareness among recreational anglers	The study confirmed low reporting rates and proposed measures to increase engagement among fishery participants.
“Electronic Self-reporting: Angler Attitudes and Behaviors in the Recreational Red Snapper Fishery” [17]	Behavioral and attitudinal analysis of self-reporting	User acceptance was found to be crucial, with reliability and convenience significantly influencing participation rates.

, previous studies on EM and ER (electronic reporting) have focused on the technical effectiveness of one system alone, which limits consideration of user acceptance and the potential of combined system integration. For example, EM trials in the western Pacific tuna longline fishery demonstrated higher species diversity detection compared to logbook records, underscoring the value of objective, video-based evidence [14]. In the case of Egypt’s Mediterranean fisheries, a low-cost digital workflow was proposed to address the challenges faced by low- and middle-income countries in establishing MCS systems [15]. Despite these insights, few studies have proposed an integrated architecture that combines EM’s objectivity with ER’s timeliness while embedding VMS for positional verification.

While being utilized by individual countries, RFMOs are still being discussed concerning specific implementation guidelines and system specifications for its adoption. Once EM is introduced, it will require not only enhanced performance but also more robust tools to combat IUU fishing.

While EM and ER are both essential components of modern fisheries monitoring, they differ fundamentally in their data sources, methods of collection, and verification capabilities. EM (electronic monitoring) relies on sensor- and video-based observation, providing objective, verifiable records of fishing activities, gear usage, and catch handling. This enables independent cross-checking against other data sources such as VMS and scientific surveys. In contrast, ER (electronic reporting) is based on self-reported logbook entries, which can deliver timely operational data but may be subject to reporting bias or incomplete records. EM requires higher initial investment in hardware and data processing but offers greater evidentiary value, whereas ER is generally less costly to implement and can be rapidly deployed across fleets. Integrating the two systems allows the strengths of each to offset the other’s limitations—combining ER’s timeliness with EM’s objectivity—to create a more comprehensive and reliable monitoring framework.

2.2. Methodology

First, the current operational status by country was researched and analyzed using baseline data on EM and ER from the literature and official websites. Second, this study examined the annual trends in IUU vessel listings from 2002 to 2024 and analyzed the composition by vessel type and fishing gear type from 2006 to 2024. The reason for commencing the vessel type and gear type analysis from 2006 is that only one vessel was listed in 2002 when RFMO IUU vessel listings began, and there was a temporary surge in listings in 2004–2005, when RFMOs began registering vessels in earnest, which does not reliably reflect recent trends. The data were obtained from Trygg Mat Tracking (TMT, Oslo, Norway), which compiles information published by RFMOs, excluding cross-listed vessels. Based on the analysis results, we considered whether there have been changes in the composition and trends despite the significant efforts made to eradicate IUU fishing on the high seas. Finally, this study adopts a concept proposal approach to develop the I-EMR model and its implementation strategy, drawing on global EMS/ERS operational cases, RFMO policy frameworks, and AI-enabled monitoring practices. The methodology involved mapping functional requirements of EM, ER, and VMS into an integrated architecture, defining data integration points for real-time cross-verification, specifying onboard AI modules for automated gear and species recognition, and embedding data governance measures to ensure interoperability and compliance. The following section outlines the current status of IUU activities and the use of EMS/ERS as the operational context for the proposed I-EMR model.

3. IUU and EMS/ERS Status

3.1. Current Status of IUU

3.1.1. Annual Status and Trends of IUU Vessels

Figure 1 illustrates the changes in the number of IUU vessels from 2002, when RFMOs began registering IUU vessel lists, to 2024. Notable peaks are observed in 2004–2006 and 2017–2019. The increase in 2004–2006 is presumed to be due to a temporary surge in interest and the listing of vessels that had long been discussed in connection with IUU fishing during the initial stage of RFMO registrations. Since the 1990s, IUU fishing has received increasing attention at both the international and national levels, and the term “IUU fishing” was first officially used by the Commission for the Conservation of Antarctic Marine Living Resources (CCAMLR) in 1997 [18], supporting this explanation. The rise in 2017–2019 is likely attributable to a combination of policy and technological factors. During this period, the cross-listing system for IUU vessels was expanded among RFMOs, and monitoring capacity using vessel monitoring systems (VMS) was enhanced [19,20]. In addition, the FAO introduced the Global Record (GR) in 2017 to improve vessel identification capacity [21], which also contributed significantly. Over the entire period, the average number of IUU vessels registered annually from 2002 to 2024 was 9.52, with 12 vessels registered in 2023 and 14 in 2024, both above the average. These results indicate that while strengthening various policy and technological capabilities has aided in identifying vessels engaged in IUU fishing, IUU fishing itself has also become more sophisticated. Therefore, monitoring, control, and surveillance (MCS) technologies must continue to evolve.

3.1.2. Status and Trends of IUU Vessels by Vessel Type

By vessel type, fishing vessels were the most prevalent (Figure 2), which is an expected result given that fishing vessels account for the largest share of the total fleet. An unusual finding is that in 2014–2015, there was a temporary increase in the number of unknown vessels; however, the improvement in vessel identification capabilities since 2017 is likely to have contributed to the subsequent decrease in unknown vessels. Although the trend for reefers shows little variation, considering that a single reefer can transport the catch from many fishing vessels, it is essential to enhance institutional and technological MCS capabilities for reefers in the same way as for fishing vessels.

3.1.3. Status and Trends of IUU Vessels by Fishing Gear Type

Figure 3 presents the classification of fishing gear types for vessels identified as fishing vessels in Figure 3. The increase observed in 2017–2019 is attributed to the same causes as the annual trend; therefore, this section focuses on changes in the composition by fishing gear type. According to Figure 3, since 2017, the ability to identify whether a vessel is a reefer or a fishing vessel appears to have improved; however, the capability to identify the fishing gear type of a fishing vessel does not seem to have advanced. The most reliable method for determining the fishing gear type of a vessel that is not flagged to the inspecting state is for an authorized inspector to physically identify the vessel at sea or in port. Information obtained through such inspections can be shared using the PSMA Global Record (GR) or the Global Information Exchange System (GIES); however, these methods are constrained by limited personnel and facilities as well as restricted access to information from non-member states. The next most commonly used approach is to utilize information registered in the Automatic Identification System (AIS). However, aside from positional data, vessel identification information in the AIS is self-reported and can be modified at any time, meaning it may be inaccurate or absent altogether. Ultimately, information managed by the flag state, such as that from VMS or electronic reporting (ER), is the most comprehensive and reliable.

Although the MCS framework has advanced significantly through extensive international efforts, IUU fishing has simultaneously evolved into a more sophisticated activity through covert and adaptive techniques. The number of IUU vessels is rebounding, and difficulties in identifying fishing gear types are increasing. While electronic reporting (ER) can provide accurate identification of fishing gear types, the information is self-reported and therefore cannot be fully trusted.

Accordingly, this study proposes the integrated electronic monitoring and reporting (I-EMR) model, which combines EM and ER functionalities on the basis of existing MCS technology—namely VMS—with the expectation that it can address these existing limitations. The I-EMR is based on accurate vessel information, integrates existing MCS tools, and applies artificial intelligence (AI) technology to comprehensively and automatically monitor all fishing activities, thereby enhancing reliability. ER has already been adopted and implemented in many countries. In contrast, EM remains in its early stages of introduction and is currently under active discussion within RFMOs. Therefore, it is necessary to examine the existing systems.

3.2. Current Status and Problems of EMS and ERS

3.2.1. Current Status of EMS and Problems of the Traditional System

Electronic monitoring systems (EMS) typically consists of multiple activity sensors, a global positioning system (GPS), computer hardware, and cameras (Figure 4), enabling video monitoring and documentation [22].

The current status of EM adoption and operation by major countries as well as the applied technologies is summarized in

Table 2. Overview of EM Implementation and Operation by Country.

Country	Target Fisheries	Monitoring Scope	Applied Technologies and Features
Australia	Trawl, longline, handline, trap, set nets	Protected species bycatch, gear deployment, catch documentation	High-resolution cameras, real-time monitoring, full coverage
Canada	Trap, longline, groundfish trawl	Catch monitoring, gear use	Cameras, sensors, GPS, continuous EM system operation
EU	Bottom trawl, set net, longline, purse seine	Full catch documentation, protected species, compliance behavior	AI trials for species identification, basket-view cameras, limited 4G transmission
New Zealand	Trawl, set nets	Bycatch handling, gear use, catch quantity	Onboard video monitoring, precision video recording, in-trawl cameras
USA	Trawl, longline, trap	Catch, effort, protected species bycatch, compliance	R&D on automated video analysis, random video sampling

based on prior research [7]. Although similar fishing methods such as trawling, longlining, potting, and purse seining were used across regions, technological differences were evident. For example, Australia and the EU (e.g., Denmark) had systems capable of transmitting EM data in real time, while other countries employed systems that stored video for later analysis.

Table 3. EM Implementation Status.

Country	Target Fisheries	Monitoring Scope	Applied Technologies and Features
China [23]	Distant-water fisheries	Compliance, catch documentation, IUU prevention	EMS with cameras and GPS, cloud-based data storage, AI-assisted video review
Japan [24]	Tuna longline, coastal, and distant-water fisheries	Catch, bycatch, CPUE	EM complements human observers; emphasis on transparency
Republic of Korea (Internal data)	Distant-water tuna longline	Catch data collection, fishing activity monitoring, preparation for international compliance	AI-based video recognition and catch estimation; EMS development ongoing (2025–2028 R&D project)

presents the current status of EM operations among large-scale fishing nations in the Pacific. These countries differ from those listed in Table 2, as they primarily target vessels operating within their own exclusive economic zones (EEZ). The analysis revealed that high-seas tuna longline fishing was a shared characteristic, and the use of cameras, sensors and GPS was common across all countries. However, differences emerged regarding efforts to incorporate AI technology. The Republic of Korea has been promoting an EM pilot project since 2022 and is advancing EMS development through an AI-based R&D initiative scheduled for 2025–2028.

While Table 3 summarizes EM implementation among major Pacific fishing nations,

Table 4. Comparison of the current statuses of EM introduction between WCPFC and IATTC.

Items	WCPFC	IATTC
Start of discussion	2014	2019
Organizational structure for discussions	Operation of EM/ER Working Group (W/G), etc.	Operation of Scientific Advisory Committee and Working Group (W/G), etc.
CMM/ Standard draft	Drafting of EM CMM and minimum standards [25]; adopting the interim EM technical standards (2024)	Drafting of EM standard [26] and roadmap (2021–2025); adopting a resolution on the interim EM minimum standards [27] (2024)
Objects of application	Western and Central Pacific longline and purse seine fisheries	Eastern Pacific tuna longline and purse seine fisheries
Main content	The main content includes the operation of the EM/ER Working Group, the establishment of minimum standards for the EM program, and the definition of the EM system together with data ownership regulations	The roadmap for EM minimum standards [26] (2021–2025), the pilot project for longline and purse seine, and the four key components of the EM standard: definitions, institutional structure, management guidelines, and technical standards
Implementation roadmap	Discussions with member countries following EMP standard [25] establishment; exploring EM applications in 2025	Review the interim standards [27], aiming for full implementation in 2027

shifts the focus to RFMO-level initiatives, particularly the WCPFC and IATTC, highlighting differences in regulatory approaches and timelines. The Pacific RFMO, WCPFC, and IATTC have established conservation and management measures (CMMs) as regulatory standards. The current statuses of EM adoption are compared in Table 4 based on previous studies [10] and the 2024 annual meeting reports of WCPFC and IATTC [25,26,27].

Based on the review so far, a major advantage of EM is that it operates using independent reporting and image-based records, making it highly effective in detecting IUU fishing and even collecting supporting evidence. Additionally, as species identification accuracy improves through iterative retraining with labeled images, overall detection accuracy increases, enhancing the system’s potential for continuous development.

However, operating EMS also presents several limitations. First, each fishing vessel requires the installation of at least two cameras, and continuous 24/7 image capture places considerable strain on onboard data storage capacity. Because images are stored for extended periods before land-based analysis, detecting IUU fishing in real time is challenging. While satellite transmission of EM data could address this, it would incur substantial communication and monitoring costs. In other words, increased operational costs and lower user acceptance present practical barriers to widespread EMS adoption.

3.2.2. Current Status of ERS and Problems of the Traditional System

Electronic reporting systems (ERS) are used by various countries to record, report, process, store, and transmit fisheries data, including information on catch, landings, sales, and transshipment. A central component is the electronic logbook, in which the vessel master records fishing operations. These records are then submitted to national authorities, which store the information in a secure database [28].

Table 5. Key features of ERS operation cases.

Country	Key Features
EU [29]	Expanded the ER of catch data and implemented automated cross-validation of fishery data
Republic of Korea (Internal data)	Pilot EM/ER integration project on distant-water fishing vessels; use of digital logbooks and a cross-checking system
United States [30]	Digital logbooks for commercial and recreational fisheries; integration with observer programs; near-real time data submission

presents the key functions used by countries that have adopted ERS, with most systems demonstrating a similar level of capability.

Next, we specifically examined the Republic of Korea’s ERS system. The ERS operated by the Fisheries Management Center (FMC) is satellite-based and is used to monitor Republic of Korea-flagged distant-water fishing and support vessels. A reporting application is installed on board each vessel, and the FMC has established a system capable of recording, analyzing, and managing the data transmitted from these vessels.

As shown in Figure 5, distant-water fishing vessels can report all information related to fishing and navigation, including departure and arrival notifications, fishing activity details (e.g., setting, hauling, etc.), species caught, catch volume, transshipment, and landing reports. Additionally, bidirectional communication between the FMC and fishing vessels enables preventive measures to be taken in advance when safety concerns or potential IUU fishing arise. Limited information can also be shared with relevant authorities in real time.

Based on the current system review, a key advantage of ER is its lower cost and ease of dissemination compared to EM. However, because it relies on self-reporting, there is a risk of intentional data omission and unintentional human error, such as incorrect entries or missing data. In practice, ER data can be used to track inconsistencies in reported species and catch volumes across fishing, transshipment and landing stages, potentially indicating IUU fishing. Nonetheless, it remains difficult to obtain evidence directly from the ER alone.

4. Proposed I-EMR Model and Framework

4.1. I-EMR Model and Framework

Integrating EM and ER systems using AI enhances the ability to prevent IUU fishing compared to operating them independently while also reducing operational costs. Specifically, eliminating the need to manage separate EM and ER systems can lower implementation expenses, thereby improving user acceptance and reducing reluctance to adopt the system.

The AI-driven I-EMR model integrates video analysis results from EM with information on vessel position, shooting and line-hauling activities (including time, gear type, and gear quantity), species identification, and video records. As this integrated dataset accumulates over time, algorithmic analysis enables the system to autonomously detect and classify fishing activities. Furthermore, when this auto-recognized data are matched with the vessel monitoring systems (VMS) for verification, they can be transmitted in real time to national FMCs and relevant fisheries agencies—without human intervention or risk of operator error. Figure 6 presents the conceptual framework for this AI-integrated I-EMR model.

Next, the procedure from recognition to analysis and reporting of the I-EMR system proposed in Figure 6 is presented. In the proposed I-EMR architecture, the integration of EM, ER, and VMS data occurs at two levels. Onboard the vessel, a local gateway device collects EM video/sensor outputs and ER electronic logbook entries in near-real time. These streams are time-stamped and geo-referenced before being merged into a unified data package. All fishing-related information is transmitted to the FMC. However, due to the limitations of satellite-based communication, it is not feasible to transmit all EM video data in real time. When an AI-based onboard system detects potential issues, the relevant video segments should be captured and transmitted, while the complete video footage should be downloaded at the landing site. At the Fisheries Monitoring Center (FMC), incoming integrated data are stored in a centralized database, where they are cross-verified against VMS position reports. This dual-level integration allows certain automated analyses—such as species identification, gear detection, and catch estimation—to be performed locally for rapid alerts, while more computationally intensive processes, including long-term pattern analysis and anomaly detection, are carried out at the FMC. Furthermore, data related to scientific collection should be made available for transfer to national research institutions. The VMS feed is continuously linked to both EM and ER datasets through a shared vessel identifier and synchronized timestamps, enabling cross-checks between reported activities, observed footage, and positional data. This ensures that discrepancies (e.g., mismatches between logbook entries and recorded video) can be detected and flagged immediately.

Table 6. Components supporting stable operation of the I-EMR Model.

Process Stage	Function	EM	ER
Data Collection	Video recording during fishing (standardized resolution, GPS watermark)	●	-
	Capture electronic logbook entries (permits, reports, fishing info)	-	●
	Collect scientific data	●	●
Data Analysis	Gear recognition	●	-
	Species classification	●	-
	IUU fishing recognition	●	-
	Fishing activity determination (based on EM video or ER logbook)	●	●
Machine Learning	Data preprocessing and labeling	●	●
	Data augmentation (diversification of training data)	●	●
	AI model creation (e.g., YOLO Vx)	●
Integration and Communication	Merge EM and ER data streams; cross-check with VMS	●	●
	Real-time data/video transmission and compression	●
	Bidirectional information exchange	●	●
Information Management	Secure data storage and encryption	●	●
	Centralized integrated database	●	●
	Access control and user permission management	●	●
	Object recognition and analysis servers	●
	Web/streaming/search server	●

presents the key equipment components required to ensure the stable operation of the I-EMR model outlined in Figure 7. The figure outlines subsystems for video acquisition, analysis, learning, and information management mapped to EM, ER, and VMS roles. These components provide the technical foundation for a robust and scalable I-EMR system aligned with international fisheries monitoring standards.

4.2. Measures to Develop Effective I-EMR System

The following measures are essential for establishing an effective and stable I-EMR system.

First, an automated system must be developed to classify and store only video footage relevant to fishing activities or suspected IUU fishing incidents using AI algorithms. This will minimize the volume of EM and ER data retained. The system should also be designed to regularly refine incoming data, retaining only essential information. Additionally, simplifying the system’s user interface (UI) will help reduce physical storage demands by avoiding the accumulation of redundant or long-term data.

Second, collaborative participation by RFMOs in establishing standardized system designs will reduce initial implementation costs and mitigate issues such as inconsistent system configurations and development delays. Standardization is critical, as vessels often operate across multiple RFMO jurisdictions. To facilitate this, a dedicated I-EMR working group or technical committee—coordinated under the FAO—could be formed to define shared technical specifications and operational guidelines.

Third, as a technical and institutional safeguard, it is essential to encrypt data and assign user access rights on a per-user basis to protect vessel and personal information. Institutionally, sensitive data should be minimized and, in unavoidable cases, managed internally on a public-interest basis in accordance with RFMO Conservation and CMMs and national regulations, ensuring no unauthorized external disclosure.

Also, to ensure practical and stable adoption, the development of the I-EMR system should follow a phased roadmap; accordingly, a three-phase implementation strategy is presented, as shown in Figure 7.

In Phase 1, the strategy focuses on targeted pilot projects in priority fisheries. At this stage, EM, ER, and VMS are integrated within a limited fleet segment in order to refine both the system design and the AI algorithms.

In Phase 2, the implementation is expanded to additional fisheries. This phase is supported by technical capacity building, the establishment of standardized operating procedures, and enhanced coordination across RFMOs.

In Phase 3, the system advances toward full-scale deployment. It is embedded into national and RFMO compliance frameworks, while data-sharing protocols are established to support scientific research.

This staged approach minimizes operational risks, aligns infrastructure requirements with available resources, and facilitates long-term sustainability through gradual scaling.

4.3. Discussion and Limitations

First, EM systems offer greater consistency and cost efficiency than human observers by minimizing review time and reducing data redundancy [13]. In the long term, the integration of AI-based video analysis and optimized program design through EM and ER can further enhance cost effectiveness [13]. Fully implemented programs are often driven by strong compliance or management issues, such as gear theft or high discard rates—as seen in the British Columbia “Area A” crab fishery program—where EM proved to be the most cost-effective solution [31]. The I-EMR model proposed in this study, based on AI integration of EM and ER, enables more efficient fisheries management while reducing operational costs.

Second, video-based monitoring systems improve traceability of catches, deter the concealment of bycatch and discards, and enhance the accuracy of fishery-related reporting [8]. The primary goal of EM is monitoring, and its effectiveness has been demonstrated in various gear types, including longline, purse seine, trawl, gillnet, and pot/trap methods [32]. The I-EMR model further strengthens these outcomes by linking catch data with vessel behavior and fishing gear information.

Third, the real-time analysis capabilities of the I-EMR system can support MCS policies. Recent MCS frameworks emphasize a fit-for-purpose approach that allocates resources based on risk levels associated with fishing grounds or vessels [33]. I-EMR enables flag states to selectively monitor their vessels and supports RFMOs in optimizing surveillance efforts around high-risk targets.

Finally, the integration of EM and ER enables data-driven decision making for fisheries resource management. Catch data obtained through EM provide a fuller picture of fishing activities [34], support stock assessments, and facilitate the sustainable use of catch limits [35]. In other words, high-quality data generated through I-EMR can be used for resource evaluation, compliance verification, and policy effectiveness analysis. I-EMR provides real-time indicators on catch volume, fishing effort, and the likelihood of IUU activities, thus supporting science-based, adaptive management frameworks.

In the proposed I-EMR model, the AI workflow follows a structured, sequential process with feedback loops for continuous improvement. First, raw data are collected from EM cameras and sensors (e.g., video footage of fishing operations, gear deployment, and catch handling) and from ER systems (e.g., electronic logbook entries on catch, effort, and permits). These inputs are automatically time-stamped and geo-referenced. Second, preprocessing module filters and segments the video into relevant fishing events, applying AI algorithms for object detection (gear type, species, and catch volume) and behavioral analysis (e.g., hauling, setting, and transshipment). Third, the extracted features are cross-verified in near-real time with positional data from the VMS to ensure accuracy. Fourth, validated datasets are transmitted to the Fisheries Monitoring Center (FMC), where more computationally intensive analytics—such as anomaly detection, compliance scoring, and long-term pattern analysis—are performed. Finally, the AI models are retrained using labeled datasets from verified cases, enabling continuous learning and incremental improvement in detection accuracy. This explicit workflow ensures that EM, ER, and VMS data are fully integrated, systematically analyzed, and used to support both real-time monitoring and long-term fisheries management objectives.

While the proposed I-EMR framework offers a conceptual pathway toward integrating EM, ER, and VMS for more effective MCS, it is important to acknowledge that the model remains hypothetical and has not undergone empirical validation through pilot testing or simulation. As such, the performance metrics—including detection accuracy, processing latency, and cost–benefit efficiency—are yet to be quantified. Practical implementation also faces potential challenges: (1) data acquisition constraints, such as inconsistent EM coverage and limited ER adoption in some fleets; (2) computational demands, particularly for onboard AI processing of high-resolution video streams; and (3) legal and institutional barriers, including data-sharing restrictions between RFMOs and national authorities. Future research should therefore prioritize controlled trials of the I-EMR system across diverse fisheries to assess its operational feasibility, validate AI-based detection algorithms, and evaluate cost effectiveness relative to existing MCS methods. Additionally, cross-jurisdictional collaboration will be essential to harmonize system specifications, resolve data governance issues, and facilitate interoperability between RFMO and national monitoring frameworks.

5. Conclusions

This study presented a conceptual framework for the intelligent electronic monitoring and reporting (I-EMR) model, integrating EM, ER, and VMS to enhance monitoring accuracy, data reliability, and cost efficiency in combating IUU fishing. Key findings include the feasibility of minimizing redundant data through AI-driven video selection, the importance of standardized system design across RFMOs, and the necessity of robust data security protocols to increase user acceptance.

From a policy perspective, the I-EMR model offers a scalable solution for improving compliance with conservation and management measures (CMMs), strengthening RFMO cooperation and supporting adaptive fisheries management through high-quality, real-time data. Its ability to integrate multiple monitoring streams into a unified platform can facilitate transparent governance, enable targeted surveillance of high-risk vessels, and inform science-based decision making at both national and international levels. In addition, the ability of the I-EMR system to monitor onboard activities can serve as an important tool for improving vessel safety, as it enables both real-time awareness of safety incidents and subsequent accident investigation.

While this study is conceptual and lacks empirical validation, future research should include pilot testing in diverse fisheries to assess technical performance, institutional feasibility, and cost effectiveness. Continued RFMO discussions will be critical to harmonizing standards, refining AI detection algorithms, and ensuring that the I-EMR system can serve as a globally recognized tool for sustainable fisheries governance.