Special Issue "Artificial Intelligence and Big Data Computing"

A special issue of Mathematics (ISSN 2227-7390). This special issue belongs to the section "Engineering Mathematics".

Deadline for manuscript submissions: closed (31 May 2021) | Viewed by 29081

Special Issue Editor

Prof. Dr. Bo-Hao Chen
E-Mail Website
Guest Editor
Department of Computer Science and Engineering, Yuan Ze University, Taoyuan 32003, Taiwan
Interests: computer vision; deep learning; autonomous vehicles; knowledge representation/reasoning; human-computer interaction

Special Issue Information

Dear Colleagues,

With recent advances in information technology, the amount of digital data in several areas, ranging from business intelligence to scientific explorations, has exploded in recent years. The massive volume of both structured and unstructured data will present challenges as well as opportunities for industries and academia over the next few years.

Artificial intelligence has become one of the core areas in big data computing to offer adaptive mechanisms that enable the understanding of voluminous data in complex and changing environments. The fundamental building blocks of artificial intelligence and big data computing involve computational modeling of computer vision systems, biologically intelligent systems, multi-agent systems, cyber-physical systems, information security systems, etc.

We invite our colleagues to submit papers related to both aspects of artificial intelligence and big data computing. Artificial intelligence includes (but is not limited to) search, planning, knowledge representation, reasoning, natural language processing, robotics and perception, multi-agent systems, vision systems, statistical learning, reinforcement learning, and deep learning. Big data computing includes topics that apply artificial intelligence to data visualization, information retrieval, graph mining, cloud and grid computing, parallel and distributed computing, cryptography, blockchain, cryptocurrency, and bioinformatics.

Prof. Dr. Bo-Hao Chen
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Mathematics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2100 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • Artificial neural networks and learning methods
  • Knowledge-based systems
  • Pattern recognition
  • Agent models and architectures
  • Cloud computing techniques
  • Mobile and pervasive computing
  • Graph mining
  • Visualization
  • Blockchain and cryptocurrency
  • Biomedical informatics

Published Papers (14 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

Article
A Novel Approach Combining Particle Swarm Optimization and Deep Learning for Flash Flood Detection from Satellite Images
Mathematics 2021, 9(22), 2846; https://doi.org/10.3390/math9222846 - 10 Nov 2021
Cited by 5 | Viewed by 1651
Abstract
Flood is one of the deadliest natural hazards worldwide, with the population affected being more than 2 billion between 1998–2017 with a lack of warning systems according to WHO. Especially, flash floods have the potential to generate fatal damages due to their rapid [...] Read more.
Flood is one of the deadliest natural hazards worldwide, with the population affected being more than 2 billion between 1998–2017 with a lack of warning systems according to WHO. Especially, flash floods have the potential to generate fatal damages due to their rapid evolution and the limited warning and response time. An effective Early Warning Systems (EWS) could support detection and recognition of flash floods. Information about a flash flood can be mainly provided from observations of hydrology and from satellite images taken before the flash flood happens. Then, predictions from satellite images can be integrated with predictions based on sensors’ information to improve the accuracy of a forecasting system and subsequently trigger warning systems. The existing Deep Learning models such as UNET has been effectively used to segment the flash flood with high performance, but there are no ways to determine the most suitable model architecture with the proper number of layers showing the best performance in the task. In this paper, we propose a novel Deep Learning architecture, namely PSO-UNET, which combines Particle Swarm Optimization (PSO) with UNET to seek the best number of layers and the parameters of layers in the UNET based architecture; thereby improving the performance of flash flood segmentation from satellite images. Since the original UNET has a symmetrical architecture, the evolutionary computation is performed by paying attention to the contracting path and the expanding path is synchronized with the following layers in the contracting path. The UNET convolutional process is performed four times. Indeed, we consider each process as a block of the convolution having two convolutional layers in the original architecture. Training of inputs and hyper-parameters is performed by executing the PSO algorithm. In practice, the value of Dice Coefficient of our proposed model exceeds 79.75% (8.59% higher than that of the original UNET model). Experimental results on various satellite images prove the advantages and superiority of the PSO-UNET approach. Full article
(This article belongs to the Special Issue Artificial Intelligence and Big Data Computing)
Show Figures

Figure 1

Article
3D-DCDAE: Unsupervised Music Latent Representations Learning Method Based on a Deep 3D Convolutional Denoising Autoencoder for Music Genre Classification
Mathematics 2021, 9(18), 2274; https://doi.org/10.3390/math9182274 - 16 Sep 2021
Cited by 2 | Viewed by 1107
Abstract
With unlabeled music data widely available, it is necessary to build an unsupervised latent music representation extractor to improve the performance of classification models. This paper proposes an unsupervised latent music representation learning method based on a deep 3D convolutional denoising autoencoder (3D-DCDAE) [...] Read more.
With unlabeled music data widely available, it is necessary to build an unsupervised latent music representation extractor to improve the performance of classification models. This paper proposes an unsupervised latent music representation learning method based on a deep 3D convolutional denoising autoencoder (3D-DCDAE) for music genre classification, which aims to learn common representations from a large amount of unlabeled data to improve the performance of music genre classification. Specifically, unlabeled MIDI files are applied to 3D-DCDAE to extract latent representations by denoising and reconstructing input data. Next, a decoder is utilized to assist the 3D-DCDAE in training. After 3D-DCDAE training, the decoder is replaced by a multilayer perceptron (MLP) classifier for music genre classification. Through the unsupervised latent representations learning method, unlabeled data can be applied to classification tasks so that the problem of limiting classification performance due to insufficient labeled data can be solved. In addition, the unsupervised 3D-DCDAE can consider the musicological structure to expand the understanding of the music field and improve performance in music genre classification. In the experiments, which utilized the Lakh MIDI dataset, a large amount of unlabeled data was utilized to train the 3D-DCDAE, obtaining a denoising and reconstruction accuracy of approximately 98%. A small amount of labeled data was utilized for training a classification model consisting of the trained 3D-DCDAE and the MLP classifier, which achieved a classification accuracy of approximately 88%. The experimental results show that the model achieves state-of-the-art performance and significantly outperforms other methods for music genre classification with only a small amount of labeled data. Full article
(This article belongs to the Special Issue Artificial Intelligence and Big Data Computing)
Show Figures

Figure 1

Article
Severity Prediction for Bug Reports Using Multi-Aspect Features: A Deep Learning Approach
Mathematics 2021, 9(14), 1644; https://doi.org/10.3390/math9141644 - 13 Jul 2021
Cited by 4 | Viewed by 1190
Abstract
The severity of software bug reports plays an important role in maintaining software quality. Many approaches have been proposed to predict the severity of bug reports using textual information. In this research, we propose a deep learning framework called MASP that uses convolutional [...] Read more.
The severity of software bug reports plays an important role in maintaining software quality. Many approaches have been proposed to predict the severity of bug reports using textual information. In this research, we propose a deep learning framework called MASP that uses convolutional neural networks (CNN) and the content-aspect, sentiment-aspect, quality-aspect, and reporter-aspect features of bug reports to improve prediction performance. We have performed experiments on datasets collected from Eclipse and Mozilla. The results show that the MASP model outperforms the state-of-the-art CNN model in terms of average Accuracy, Precision, Recall, F1-measure, and the Matthews Correlation Coefficient (MCC) by 1.83%, 0.46%, 3.23%, 1.72%, and 6.61%, respectively. Full article
(This article belongs to the Special Issue Artificial Intelligence and Big Data Computing)
Show Figures

Figure 1

Article
An Efficient DA-Net Architecture for Lung Nodule Segmentation
Mathematics 2021, 9(13), 1457; https://doi.org/10.3390/math9131457 - 22 Jun 2021
Cited by 17 | Viewed by 1494
Abstract
A typical growth of cells inside tissue is normally known as a nodular entity. Lung nodule segmentation from computed tomography (CT) images becomes crucial for early lung cancer diagnosis. An issue that pertains to the segmentation of lung nodules is homogenous modular variants. [...] Read more.
A typical growth of cells inside tissue is normally known as a nodular entity. Lung nodule segmentation from computed tomography (CT) images becomes crucial for early lung cancer diagnosis. An issue that pertains to the segmentation of lung nodules is homogenous modular variants. The resemblance among nodules as well as among neighboring regions is very challenging to deal with. Here, we propose an end-to-end U-Net-based segmentation framework named DA-Net for efficient lung nodule segmentation. This method extracts rich features by integrating compactly and densely linked rich convolutional blocks merged with Atrous convolutions blocks to broaden the view of filters without dropping loss and coverage data. We first extract the lung’s ROI images from the whole CT scan slices using standard image processing operations and k-means clustering. This reduces the search space of the model to only lungs where the nodules are present instead of the whole CT scan slice. The evaluation of the suggested model was performed through utilizing the LIDC-IDRI dataset. According to the results, we found that DA-Net showed good performance, achieving an 81% Dice score value and 71.6% IOU score. Full article
(This article belongs to the Special Issue Artificial Intelligence and Big Data Computing)
Show Figures

Figure 1

Article
Image Denoising Using Adaptive and Overlapped Average Filtering and Mixed-Pooling Attention Refinement Networks
Mathematics 2021, 9(10), 1130; https://doi.org/10.3390/math9101130 - 17 May 2021
Cited by 3 | Viewed by 1087
Abstract
Cameras are essential parts of portable devices, such as smartphones and tablets. Most people have a smartphone and can take pictures anywhere and anytime to record their lives. However, these pictures captured by cameras may suffer from noise contamination, causing issues for subsequent [...] Read more.
Cameras are essential parts of portable devices, such as smartphones and tablets. Most people have a smartphone and can take pictures anywhere and anytime to record their lives. However, these pictures captured by cameras may suffer from noise contamination, causing issues for subsequent image analysis, such as image recognition, object tracking, and classification of an object in the image. This paper develops an effective combinational denoising framework based on the proposed Adaptive and Overlapped Average Filtering (AOAF) and Mixed-pooling Attention Refinement Networks (MARNs). First, we apply AOAF to the noisy input image to obtain a preliminarily denoised result, where noisy pixels are removed and recovered. Next, MARNs take the preliminary result as the input and output a refined image where details and edges are better reconstructed. The experimental results demonstrate that our method performs favorably against state-of-the-art denoising methods. Full article
(This article belongs to the Special Issue Artificial Intelligence and Big Data Computing)
Show Figures

Figure 1

Article
Subsurface Temperature Estimation from Sea Surface Data Using Neural Network Models in the Western Pacific Ocean
Mathematics 2021, 9(8), 852; https://doi.org/10.3390/math9080852 - 14 Apr 2021
Cited by 8 | Viewed by 1367
Abstract
Estimating the ocean subsurface thermal structure (OSTS) based on multisource sea surface data in the western Pacific Ocean is of great significance for studying ocean dynamics and El Niño phenomenon, but it is challenging to accurately estimate the OSTS from sea surface parameters [...] Read more.
Estimating the ocean subsurface thermal structure (OSTS) based on multisource sea surface data in the western Pacific Ocean is of great significance for studying ocean dynamics and El Niño phenomenon, but it is challenging to accurately estimate the OSTS from sea surface parameters in the area. This paper proposed an improved neural network model to estimate the OSTS from 0–2000 m from multisource sea surface data including sea surface temperature (SST), sea surface salinity (SSS), sea surface height (SSH), and sea surface wind (SSW). In the model experiment, the rasterized monthly average data from 2005–2015 and 2016 were selected as the training and testing set, respectively. The results showed that the sea surface parameters selected in the paper had a positive effect on the estimation process, and the average RMSE value of the ocean subsurface temperature (OST) estimated by the proposed model was 0.55 °C. Moreover, there were pronounced seasonal variation signals in the upper layers (the upper 200 m), however, this signal gradually diminished with increasing depth. Compared with known estimation models such as the random forest (RF), the multiple linear regression (MLR), and the extreme gradient boosting (XGBoost), the proposed model outperformed these models under the data conditions of the paper. This research can provide an advanced artificial intelligence technique for estimating subsurface thermohaline structure in major sea areas. Full article
(This article belongs to the Special Issue Artificial Intelligence and Big Data Computing)
Show Figures

Figure 1

Article
Data-Driven Leak Localization in Urban Water Distribution Networks Using Big Data for Random Forest Classifier
Mathematics 2021, 9(6), 672; https://doi.org/10.3390/math9060672 - 22 Mar 2021
Cited by 7 | Viewed by 1963
Abstract
In the present paper, a Random Forest classifier is used to detect leak locations on two different sized water distribution networks with sparse sensor placement. A great number of leak scenarios were simulated with Monte Carlo determined leak parameters (leak location and emitter [...] Read more.
In the present paper, a Random Forest classifier is used to detect leak locations on two different sized water distribution networks with sparse sensor placement. A great number of leak scenarios were simulated with Monte Carlo determined leak parameters (leak location and emitter coefficient). In order to account for demand variations that occur on a daily basis and to obtain a larger dataset, scenarios were simulated with random base demand increments or reductions for each network node. Classifier accuracy was assessed for different sensor layouts and numbers of sensors. Multiple prediction models were constructed for differently sized leakage and demand range variations in order to investigate model accuracy under various conditions. Results indicate that the prediction model provides the greatest accuracy for the largest leaks, with the smallest variation in base demand (62% accuracy for greater- and 82% for smaller-sized networks, for the largest considered leak size and a base demand variation of ±2.5%). However, even for small leaks and the greatest base demand variations, the prediction model provided considerable accuracy, especially when localizing the sources of leaks when the true leak node and neighbor nodes were considered (for a smaller-sized network and a base demand of variation ±20% the model accuracy increased from 44% to 89% when top five nodes with greatest probability were considered, and for a greater-sized network with a base demand variation of ±10% the accuracy increased from 36% to 77%). Full article
(This article belongs to the Special Issue Artificial Intelligence and Big Data Computing)
Show Figures

Figure 1

Article
INCO-GAN: Variable-Length Music Generation Method Based on Inception Model-Based Conditional GAN
Mathematics 2021, 9(4), 387; https://doi.org/10.3390/math9040387 - 15 Feb 2021
Cited by 9 | Viewed by 2741
Abstract
Deep learning has made significant progress in the field of automatic music generation. At present, the research on music generation via deep learning can be divided into two categories: predictive models and generative models. However, both categories have the same problems that need [...] Read more.
Deep learning has made significant progress in the field of automatic music generation. At present, the research on music generation via deep learning can be divided into two categories: predictive models and generative models. However, both categories have the same problems that need to be resolved. First, the length of the music must be determined artificially prior to generation. Second, although the convolutional neural network (CNN) is unexpectedly superior to the recurrent neural network (RNN), CNN still has several disadvantages. This paper proposes a conditional generative adversarial network approach using an inception model (INCO-GAN), which enables the generation of complete variable-length music automatically. By adding a time distribution layer that considers sequential data, CNN considers the time relationship in a manner similar to RNN. In addition, the inception model obtains richer features, which improves the quality of the generated music. In experiments conducted, the music generated by the proposed method and that by human composers were compared. High cosine similarity of up to 0.987 was achieved between the frequency vectors, indicating that the music generated by the proposed method is very similar to that created by a human composer. Full article
(This article belongs to the Special Issue Artificial Intelligence and Big Data Computing)
Show Figures

Figure 1

Article
Forecasting Spatially-Distributed Urban Traffic Volumes via Multi-Target LSTM-Based Neural Network Regressor
Mathematics 2020, 8(12), 2233; https://doi.org/10.3390/math8122233 - 17 Dec 2020
Cited by 5 | Viewed by 1437
Abstract
Monitoring the distribution of vehicles across the city is of great importance for urban traffic control. In particular, information on the number of vehicles entering and leaving a city, or moving between urban areas, gives a valuable estimate on potential bottlenecks and congestions. [...] Read more.
Monitoring the distribution of vehicles across the city is of great importance for urban traffic control. In particular, information on the number of vehicles entering and leaving a city, or moving between urban areas, gives a valuable estimate on potential bottlenecks and congestions. The possibility of predicting such flows in advance is even more beneficial, allowing for timely traffic management strategies and targeted congestion warnings. Our work is inserted in the context of short-term forecasting, aiming to predict rapid changes and sudden variations in the traffic volume, beyond the general trend. Moreover, it concurrently targets multiple locations in the city, providing an instant prediction outcome comprising the future distribution of vehicles across several urban locations. Specifically, we propose a multi-target deep learning regressor for simultaneous predictions of traffic volumes, in multiple entry and exit points among city neighborhoods. The experiment focuses on an hourly forecasting of the amount of vehicles accessing and moving between New York City neighborhoods through the Metropolitan Transportation Authority (MTA) bridges and tunnels. By leveraging a single training process for all location points, and an instant one-step volume inference for every location at each time update, our sequential modeling approach is able to grasp rapid variations in the time series and process the collective information of all entry and exit points, whose distinct predicted values are outputted at once. The multi-target model, based on long short-term memory (LSTM) recurrent neural network layers, was tested on a real-world dataset, achieving an average prediction error of 7% and demonstrating its feasibility for short-term spatially-distributed urban traffic forecasting. Full article
(This article belongs to the Special Issue Artificial Intelligence and Big Data Computing)
Show Figures

Figure 1

Article
Low-Cost Online Handwritten Symbol Recognition System in Virtual Reality Environment of Head-Mounted Display
Mathematics 2020, 8(11), 1967; https://doi.org/10.3390/math8111967 - 05 Nov 2020
Viewed by 1304
Abstract
Virtual reality is an important technology in the digital media industry, providing a whole new experience for most people. However, its manipulation method is more difficult than the traditional keyboard and mouse. In this research, we proposed a new low-cost online handwriting symbol [...] Read more.
Virtual reality is an important technology in the digital media industry, providing a whole new experience for most people. However, its manipulation method is more difficult than the traditional keyboard and mouse. In this research, we proposed a new low-cost online handwriting symbol recognition system to accurately identify symbols by user actions. The purpose was low cost processing without requiring a server. Experimental results showed that the average success rate of recognition was 99.8%. The execution time averaged a significantly low 0.03395 s. The proposed system is, respectively, highly reliable and at a low cost. This implies that the proposed system is suitable for applications in real-time environments. Full article
(This article belongs to the Special Issue Artificial Intelligence and Big Data Computing)
Show Figures

Figure 1

Article
Generating Photomosaics with QR Code Capability
Mathematics 2020, 8(9), 1613; https://doi.org/10.3390/math8091613 - 18 Sep 2020
Cited by 1 | Viewed by 3075
Abstract
A photomosaic is an image with two layers of visual information, including an overarching image that can be seen from a distance and a matrix of individual tile images when examined closely. This paper presents a method for generating photomosaics with an additional [...] Read more.
A photomosaic is an image with two layers of visual information, including an overarching image that can be seen from a distance and a matrix of individual tile images when examined closely. This paper presents a method for generating photomosaics with an additional layer of quick response code (QR code) information that can be accessed by typical QR code scanners in cell phones. The basic idea is to carefully classify the tile images in different categories and generate the photomosaic patches by referring to the properties of QR code modules. Three levels of construction methods for generating the proposed photomosaics in different image resolutions are proposed. The results show that the generated photomosaics have good visual quality and high robustness for decoding the QR code. The proposed method endows conventional photomosaics with the QR code capability. It extends photomosaics from exhibiting purely visual information to the linkage of multimedia data. Furthermore, it increases the feasibility and potential of applying photomosaics in diverse applications, such as activity promotions or commercial product advertisements. Full article
(This article belongs to the Special Issue Artificial Intelligence and Big Data Computing)
Show Figures

Figure 1

Article
MINE: Identifying Top-k Vital Nodes in Complex Networks via Maximum Influential Neighbors Expansion
Mathematics 2020, 8(9), 1449; https://doi.org/10.3390/math8091449 - 29 Aug 2020
Cited by 3 | Viewed by 1938
Abstract
Identifying vital nodes in complex networks is of paramount importance in understanding and controlling the spreading dynamics. Currently, this study is facing great challenges in dealing with big data in many real-life applications. With the deepening of the research, scholars began to realize [...] Read more.
Identifying vital nodes in complex networks is of paramount importance in understanding and controlling the spreading dynamics. Currently, this study is facing great challenges in dealing with big data in many real-life applications. With the deepening of the research, scholars began to realize that the analysis on traditional graph model is insufficient because many nodes in a multilayer network share connections among different layers. To address this problem both efficiently and effectively, a novel algorithm for identifying vital nodes in both monolayer and multilayer networks is proposed in this paper. Firstly, a node influence measure is employed to determine the initial leader of a local community. Subsequently, the community structures are revealed via the Maximum Influential Neighbors Expansion (MINE) strategy. Afterward, the communities are regarded as super-nodes for an iteratively folding process till convergence, in order to identify influencers hierarchically. Numerical experiments on 32 real-world datasets are conducted to verify the performance of the proposed algorithm, which shows superiority to the competitors. Furthermore, we apply the proposed algorithm in the graph of adjacencies derived from the maps of China and USA. The comparison and analysis of the identified provinces (or states) suggest that the proposed algorithm is feasible and reasonable on real-life applications. Full article
(This article belongs to the Special Issue Artificial Intelligence and Big Data Computing)
Show Figures

Figure 1

Article
The Real-Time Depth Estimation for an Occluded Person Based on a Single Image and OpenPose Method
Mathematics 2020, 8(8), 1333; https://doi.org/10.3390/math8081333 - 10 Aug 2020
Cited by 11 | Viewed by 2425
Abstract
In recent years, the breakthrough of neural networks and the rise of deep learning have led to the advancement of machine vision, which has been commonly used in the practical application of image recognition. Automobiles, drones, portable devices, behavior recognition, indoor positioning and [...] Read more.
In recent years, the breakthrough of neural networks and the rise of deep learning have led to the advancement of machine vision, which has been commonly used in the practical application of image recognition. Automobiles, drones, portable devices, behavior recognition, indoor positioning and many other industries also rely on the integrated application, and require the support of deep learning and machine vision. As for these technologies, there is a high demand for the accuracy related to the recognition of portraits or objects. The recognition of human figures is also a research goal that has drawn great attention in various fields. However, the portrait will be affected by various factors such as height, weight, posture, angle and whether it is covered or not, which affects the accuracy of recognition. This paper applies the application of deep learning to portraits with different poses and angles, especially the actual distance of a single lens for the shadowed portrait (depth estimation), so that it can be used for automatic control of drones in the future. Traditional methods for calculating depth using images are mainly divided into three types: one—single-lens estimation, two—lens estimation, and three—optical band estimation. In view of the fact that both the second and third categories require relatively large and expensive equipment to effectively perform distance calculations, numerous methods for calculating distance using a single lens have recently been produced. However, whether it is the use of traditional “units of distance measurement calibration”, “defocus distance measurement”, or the “three-dimensional grid space messages distance measurement method”, all of these face corresponding difficulties and problems. Additionally, they have to deal with outside disturbances and process the shadowed image. Therefore, under the new research method, OpenPose, which is proposed by Carnegie Mellon University, this paper intends to propose a depth algorithm for a single-lens occluded portrait to estimate the actual portrait distance for different poses, angles of view and obscuration. Full article
(This article belongs to the Special Issue Artificial Intelligence and Big Data Computing)
Show Figures

Figure 1

Article
Lung X-ray Segmentation using Deep Convolutional Neural Networks on Contrast-Enhanced Binarized Images
Mathematics 2020, 8(4), 545; https://doi.org/10.3390/math8040545 - 07 Apr 2020
Cited by 12 | Viewed by 4963
Abstract
Automatically locating the lung regions effectively and efficiently in digital chest X-ray (CXR) images is important in computer-aided diagnosis. In this paper, we propose an adaptive pre-processing approach for segmenting the lung regions from CXR images using convolutional neural networks-based (CNN-based) architectures. It [...] Read more.
Automatically locating the lung regions effectively and efficiently in digital chest X-ray (CXR) images is important in computer-aided diagnosis. In this paper, we propose an adaptive pre-processing approach for segmenting the lung regions from CXR images using convolutional neural networks-based (CNN-based) architectures. It is comprised of three steps. First, a contrast enhancement method specifically designed for CXR images is adopted. Second, adaptive image binarization is applied to CXR images to separate the image foreground and background. Third, CNN-based architectures are trained on the binarized images for image segmentation. The experimental results show that the proposed pre-processing approach is applicable and effective to various CNN-based architectures and can achieve comparable segmentation accuracy to that of state-of-the-art methods while greatly expediting the model training by up to 20.74 % and reducing storage space for CRX image datasets by down to 94.6 % on average. Full article
(This article belongs to the Special Issue Artificial Intelligence and Big Data Computing)
Show Figures

Figure 1

Back to TopTop