Research

Jump to: Review

17 pages, 4706 KiB

Open AccessArticle

Gender Identification of Chinese Mitten Crab Juveniles Based on Improved Faster R-CNN

by Hao Gu, Ming Chen and Dongmei Gan

Appl. Sci. 2024, 14(2), 908; https://doi.org/10.3390/app14020908 - 21 Jan 2024

Viewed by 865

The identification of gender in Chinese mitten crab juveniles is a critical prerequisite for the automatic classification of these crab juveniles. Aiming at the problem that crab juveniles are of different sizes and relatively small, with unclear male and female characteristics and complex [...] Read more.

The identification of gender in Chinese mitten crab juveniles is a critical prerequisite for the automatic classification of these crab juveniles. Aiming at the problem that crab juveniles are of different sizes and relatively small, with unclear male and female characteristics and complex background environment, an algorithm C-SwinFaster for identifying the gender of Chinese mitten crab juveniles based on improved Faster R-CNN was proposed. This algorithm introduces Swin Transformer as the backbone network and an improved Path Aggregation Feature Pyramid Network (PAFPN) in the neck to obtain multi-scale high-level semantic feature maps, thereby improving the gender recognition accuracy of Chinese mitten crab male and female juveniles. Then, a self-attention mechanism is introduced into the region of interest pooling network (ROI Pooling) to enhance the model’s attention to the classification features of male and female crab juveniles and reduce background interference on the detection results. Additionally, we introduce an improved non-maximum suppression algorithm, termed Softer-NMS. This algorithm refines the process of determining precise target candidate boxes by modulating the confidence level, thereby enhancing detection accuracy. Finally, the focal loss function is introduced to train the model, reducing the weight of simple samples during the training process, and allowing the model to focus more on samples that are difficult to distinguish. Experimental results demonstrate that the enhanced C-SwinFaster algorithm significantly improves the identification accuracy of male and female Chinese mitten crab juveniles. The mean average precision (mAP) of this algorithm reaches 98.45%, marking a 10.33 percentage point increase over the original model. This algorithm has a good effect on the gender recognition of Chinese mitten crab juveniles and can provide technical support for the automatic classification of Chinese mitten crab juveniles. Full article

(This article belongs to the Special Issue Advanced Pattern Recognition & Computer Vision)

► Show Figures

Figure 1

15 pages, 31048 KiB

Open AccessArticle

Multi-View Masked Autoencoder for General Image Representation

by Seungbin Ji, Sangkwon Han and Jongtae Rhee

Appl. Sci. 2023, 13(22), 12413; https://doi.org/10.3390/app132212413 - 16 Nov 2023

Viewed by 937

Abstract

Self-supervised learning is a method that learns general representation from unlabeled data. Masked image modeling (MIM), one of the generative self-supervised learning methods, has drawn attention for showing state-of-the-art performance on various downstream tasks, though it has shown poor linear separability resulting from [...] Read more.

Self-supervised learning is a method that learns general representation from unlabeled data. Masked image modeling (MIM), one of the generative self-supervised learning methods, has drawn attention for showing state-of-the-art performance on various downstream tasks, though it has shown poor linear separability resulting from the token-level approach. In this paper, we propose a contrastive learning-based multi-view masked autoencoder for MIM, thus exploiting an image-level approach by learning common features from two different augmented views. We strengthen the MIM by learning long-range global patterns from contrastive loss. Our framework adopts a simple encoder–decoder architecture, thus learning rich and general representations by following a simple process: (1) Two different views are generated from an input image with random masking and by contrastive loss, we can learn the semantic distance of the representations generated by an encoder. By applying a high mask ratio, of 80%, it works as strong augmentation and alleviates the representation collapse problem. (2) With reconstruction loss, the decoder learns to reconstruct an original image from the masked image. We assessed our framework through several experiments on benchmark datasets of image classification, object detection, and semantic segmentation. We achieved 84.3% in fine-tuning accuracy on ImageNet-1K classification and 76.7% in linear probing, thus exceeding previous studies and showing promising results on other downstream tasks. The experimental results demonstrate that our work can learn rich and general image representation by applying contrastive loss to masked image modeling. Full article

(This article belongs to the Special Issue Advanced Pattern Recognition & Computer Vision)

► Show Figures

Figure 1

18 pages, 5209 KiB

Open AccessArticle

The Machine Vision Dial Automatic Drawing System—Based on CAXA Secondary Development

by Ning Zhang, Fei Li and Enxu Zhang

Appl. Sci. 2023, 13(13), 7365; https://doi.org/10.3390/app13137365 - 21 Jun 2023

Cited by 1 | Viewed by 974

Abstract

Whether the pointer recognition of an SF6 (sulfur hexafluoride) density controller is correct or not is directly related to enterprise security. Therefore, a pointer recognition system that is highly efficient, precise, and secure is required. In this paper, we set up a detection [...] Read more.

Whether the pointer recognition of an SF6 (sulfur hexafluoride) density controller is correct or not is directly related to enterprise security. Therefore, a pointer recognition system that is highly efficient, precise, and secure is required. In this paper, we set up a detection platform for efficient detection of pointer readings. To detect the pointer with high accuracy, we propose a dial pointer angle detection method based on OpenCV that obtains high-precision dial pointer angle information. We then use Socket communication to transmit the dial pointer information to the secondary development program. Finally, we use the ObjectCRX development software package based on Visual Studio 2015 to quickly access the CAXA electronic drawing board graphics database, carry out secondary development of the CAXA electronic drawing board 2021, and comprehensively apply various functions to draw the overall outline of the dial. Finally, we read the dial pointer angle in order to draw the current test product dial CAD drawing. After verification, the average error rate of pointer angle recognition is 0.69°, which is much lower than the enterprise standard of 2.7°. Based on machine vision technology, we researched and designed an SF6 density controller dial automatic drawing system, and after many tests, the system runs stably and the accuracy meets the needs of enterprises. Full article

(This article belongs to the Special Issue Advanced Pattern Recognition & Computer Vision)

► Show Figures

Figure 1

28 pages, 10357 KiB

Open AccessArticle

Modeling of Walking-Gait Parameters and Walking Strategy for Quadruped Robots

by Zhaolu Li, Yumin Song, Xiaoli Zhang, Xiafu Peng and Ning Xu

Appl. Sci. 2023, 13(12), 6876; https://doi.org/10.3390/app13126876 - 06 Jun 2023

Viewed by 1523

Abstract

The inspiration for the footed robot was originally derived from biology, and it was an imitation of biological form and movement. In this paper, a bionic-robot dog is designed to reveal the motion characteristics of a quadruped robot mechanism through modeling, model kinematic [...] Read more.

The inspiration for the footed robot was originally derived from biology, and it was an imitation of biological form and movement. In this paper, a bionic-robot dog is designed to reveal the motion characteristics of a quadruped robot mechanism through modeling, model kinematic analysis, and other methods. First, the structural characteristics and movement characteristics of the developed bionic-dog model are studied. The first step is to study the physiological structure of the dog, analyze the function of the dog’s limbs, and then use a high-speed camera to capture the motion of the marked bionic-robot dog and shoot motion video of the bionic-robot dog in different motion states. The effective data of the marked points in the video are extracted using PHOTRON 1.0 software, and the extracted data are analyzed and processed in the software MATLAB R2020a, and finally the structural characteristics and motion laws of the bionic-robot dog are obtained. Then, a bionic-robot-dog experimental platform is built to conduct experiments with three planned gaits (dynamic gait, static gait, and gait transition). The experiments showed that the three gaits were consistent with the planned movements and the bionic-robot dog could perform stable fast-gait walking, slow-gait walking, and quickly complete gait transitions. All three gaits were simulated in ADAMS 2019 software, and the simulation results showed that all three gaits caused the bionic dog robot to move smoothly. Full article

(This article belongs to the Special Issue Advanced Pattern Recognition & Computer Vision)

► Show Figures

Figure 1

14 pages, 4191 KiB

Open AccessArticle

GAN Data Augmentation Methods in Rock Classification

by Gaochang Zhao, Zhao Cai, Xin Wang and Xiaohu Dang

Appl. Sci. 2023, 13(9), 5316; https://doi.org/10.3390/app13095316 - 24 Apr 2023

Cited by 3 | Viewed by 1254

Abstract

In this paper, a data augmentation method Conditional Residual Deep Convolutional Generative Adversarial Network (CRDCGAN) based on Deep Convolutional Generative Adversarial Network (DCGAN) is proposed to address the problem that the accuracy of existing image classification techniques is too low when classifying small-scale [...] Read more.

In this paper, a data augmentation method Conditional Residual Deep Convolutional Generative Adversarial Network (CRDCGAN) based on Deep Convolutional Generative Adversarial Network (DCGAN) is proposed to address the problem that the accuracy of existing image classification techniques is too low when classifying small-scale rock images. Firstly, Wasserstein distance is introduced to change the loss function, which makes the training of the network more stable; secondly, conditional information is added, and the network has the ability to generate and discriminate image data with label information; finally, the residual module is added to improve the quality of generated images. The results demonstrate that by applying CRDCGAN to the augmented rock image dataset, the accuracy of the classification model trained on this dataset is as high as 96.38%, which is 13.39% higher than that of the classification model trained on the non-augmented dataset, and 8.56% and 6.27% higher than that of the traditional dataset augmented method and the DCGAN dataset augmentation method, respectively. CRDCGAN expands the rock image dataset, which makes the rock classification model accuracy effectively improved. The data augmentation method was found to be able to change the accuracy of the classification model to a greater extent. Full article

(This article belongs to the Special Issue Advanced Pattern Recognition & Computer Vision)

► Show Figures

Figure 1

17 pages, 1876 KiB

Open AccessArticle

A Blind Image Quality Index for Synthetic and Authentic Distortions with Hierarchical Feature Fusion

by Lingbi Hu, Juan Peng, Tuoxun Zhao, Wei Yu and Bo Hu

Appl. Sci. 2023, 13(6), 3591; https://doi.org/10.3390/app13063591 - 11 Mar 2023

Cited by 1 | Viewed by 1336

Abstract

Blind Image Quality Assessment (BIQA) for synthetic and authentic distortions has attracted much attention in the community, and it is still a great challenge. The existing quality metrics are mildly consistent with subjective perception. Traditional handcrafted quality metrics can easily and directly extract [...] Read more.

Blind Image Quality Assessment (BIQA) for synthetic and authentic distortions has attracted much attention in the community, and it is still a great challenge. The existing quality metrics are mildly consistent with subjective perception. Traditional handcrafted quality metrics can easily and directly extract low-level features, which mainly account for the outline, edge, color, texture, and shape features, while ignoring the important deep semantics of the distorted image. In the field of popular deep learning, multilevel features can be acquired easily. However, most of them either use only high-level features, ignoring the shallow features, or they simply combine features at different levels, resulting in limited prediction performance. Motivated by these, this paper presents a novel BIQA for synthetic and authentic distortions with hierarchical feature fusion in a flexible vision-Transformer framework. First, multiscale features are extracted from a strong vision-Transformer backbone. Second, an effective hierarchical feature fusion module is proposed to incorporate the features at different levels progressively. To eliminate redundant information, a simple but effective attention mechanism is employed after each fusion. Third, inspired by the human visual system, local and global features from the fusion features are extracted to represent different granularity distortions. Finally, these local and global features are mapped to the final quality score. Extensive experiments on three authentic image databases and two synthetic image datasets show that the proposed method is superior to the state-of-the-art quality metrics for both single-database testing and cross-database testing. Full article

(This article belongs to the Special Issue Advanced Pattern Recognition & Computer Vision)

► Show Figures

Figure 1

23 pages, 5705 KiB

Open AccessArticle

Integrating Prior Knowledge into Attention for Ship Detection in SAR Images

by Yin Pan, Lei Ye, Yingkun Xu and Junyi Liang

Appl. Sci. 2023, 13(5), 2941; https://doi.org/10.3390/app13052941 - 24 Feb 2023

Cited by 2 | Viewed by 1235

Abstract

Although they have achieved great success in optical images, deep convolutional neural networks underperform for ship detection in SAR images because of the lack of color and textual features. In this paper, we propose our framework which integrates prior knowledge into neural networks [...] Read more.

Although they have achieved great success in optical images, deep convolutional neural networks underperform for ship detection in SAR images because of the lack of color and textual features. In this paper, we propose our framework which integrates prior knowledge into neural networks by means of the attention mechanism. Because the background of ships is mostly water surface or coast, we use clustering algorithms to generate the prior knowledge map from brightness and density features. The prior knowledge map is later resized and fused with convolutional feature maps by the attention mechanism. Our experiments demonstrate that our framework is able to improve various one-stage and two-stage object detection algorithms (Faster R-CNN, RetinaNet, SSD, and YOLOv4) on two benchmark datasets (SSDD, LS-SSDD, and HRSID). Full article

(This article belongs to the Special Issue Advanced Pattern Recognition & Computer Vision)

► Show Figures

Graphical abstract

22 pages, 6706 KiB

Open AccessArticle

Research on Human-Computer Interaction Technology of Large-Scale High-Resolution Display Wall System

by Chen Huang, Yimin Chen, Weiqin Tong, Tao Feng and Mingxing Deng

Appl. Sci. 2023, 13(1), 591; https://doi.org/10.3390/app13010591 - 01 Jan 2023

Cited by 1 | Viewed by 2120

Abstract

As an effective solution for data visualization and analysis, the large-scale high-resolution display wall system has been widely used in various scientific research fields. On the basis of investigating existing system cases and research results, this paper introduces the SHU-VAS system (60 screens, [...] Read more.

As an effective solution for data visualization and analysis, the large-scale high-resolution display wall system has been widely used in various scientific research fields. On the basis of investigating existing system cases and research results, this paper introduces the SHU-VAS system (60 screens, 120 megapixels) for data visualization and analysis. In order to improve the efficiency of human-computer interaction in large-scale high-definition display wall systems, we propose an interaction framework based on gesture and double-precision pointing technology. During the interaction process, an adaptive mode switching method based on user action features is used to switch between rough and precise control modes. In the evaluation experiments, we analyzed the characteristics of different interaction methods in movement and navigation interaction tasks, and verified the effectiveness of the gesture-based interaction framework proposed in this paper. Full article

(This article belongs to the Special Issue Advanced Pattern Recognition & Computer Vision)

► Show Figures

Figure 1

18 pages, 5484 KiB

Open AccessArticle

Multiple Pedestrian Tracking in Dense Crowds Combined with Head Tracking

by Zhouming Qi, Mian Zhou, Guoqiang Zhu and Yanbing Xue

Appl. Sci. 2023, 13(1), 440; https://doi.org/10.3390/app13010440 - 29 Dec 2022

Cited by 1 | Viewed by 1752

Abstract

In order to reduce the negative impact of severe occlusion in dense scenes on the performance degradation of the tracker, considering that the head is the highest and least occluded part of the pedestrian’s entire body, we propose a new multiobject tracking method [...] Read more.

In order to reduce the negative impact of severe occlusion in dense scenes on the performance degradation of the tracker, considering that the head is the highest and least occluded part of the pedestrian’s entire body, we propose a new multiobject tracking method for pedestrians in dense crowds combined with head tracking. For each frame of the video, a head tracker is first used to generate the pedestrians’ head movement tracklets, and the pedestrians’ whole body bounding boxes are detected at the same time. Secondly, the degree of association between the head bounding boxes and the whole body bounding boxes are calculated, and the Hungarian algorithm is used to match the above calculation results. Finally, according to the matching results, the head bounding boxes in the head tracklets are replaced with the whole body bounding boxes, and the whole body motion tracklets of the pedestrians in the dense scene are generated. Our method can be performed online, and experiments suggested that our method effectively reduces the negative effects of false negatives and false positives on the tracker caused by severe occlusion in dense scenes. Full article

(This article belongs to the Special Issue Advanced Pattern Recognition & Computer Vision)

► Show Figures

Figure 1

22 pages, 2988 KiB

Open AccessArticle

Efficient Object Detection in SAR Images Based on Computation-Aware Neural Architecture Search

by Chuanyou Li, Yifan Li, Huanyun Hu, Jiangwei Shang, Kun Zhang, Lei Qian and Kexiang Wang

Appl. Sci. 2022, 12(21), 10978; https://doi.org/10.3390/app122110978 - 29 Oct 2022

Cited by 1 | Viewed by 2036

Abstract

Remote sensing techniques are becoming more sophisticated as radar imaging techniques mature. Synthetic aperture radar (SAR) can now provide high-resolution images for day-and-night earth observation. Detecting objects in SAR images is increasingly playing a significant role in a series of applications. In this [...] Read more.

Remote sensing techniques are becoming more sophisticated as radar imaging techniques mature. Synthetic aperture radar (SAR) can now provide high-resolution images for day-and-night earth observation. Detecting objects in SAR images is increasingly playing a significant role in a series of applications. In this paper, we address an edge detection problem that applies to scenarios with ship-like objects, where the detection accuracy and efficiency must be considered together. The key to ship detection lies in feature extraction. To efficiently extract features, many existing studies have proposed lightweight neural networks by pruning well-known models in the computer vision field. We found that although different baseline models have been tailored, a large amount of computation is still required. In order to achieve a lighter neural network-based ship detector, we propose Darts_Tiny, a novel differentiable neural architecture search model, to design dedicated convolutional neural networks automatically. Darts_Tiny is customized from Darts. It prunes superfluous operations to simplify the search model and adopts a computation-aware search process to enhance the detection efficiency. The computation-aware search process not only integrates a scheme cutting down the number of channels on purpose but also adopts a synthetic loss function combining the cross-entropy loss and the amount of computation. Comprehensive experiments are conducted to evaluate Darts_Tiny on two open datasets, HRSID and SSDD. Experimental results demonstrate that our neural networks win by at least an order of magnitude in terms of model complexity compared with SOTA lightweight models. A representative model obtained from Darts_Tiny (158 KB model volume, 28 K parameters and

0.58

G computations) yields a faster detection speed such that more than 750 frames per second (

800 \times 800

SAR images) could be achieved when testing on a platform equipped with an Nvidia Tesla V100 and an Intel Xeon Platinum 8260. The lightweight neural networks generated by Darts_Tiny are still competitive in detection accuracy: the F1 score can still reach more than 83 and 90, respectively, on HRSID and SSDD. Full article

(This article belongs to the Special Issue Advanced Pattern Recognition & Computer Vision)

► Show Figures

Figure 1

15 pages, 3853 KiB

Open AccessArticle

A New Knowledge-Distillation-Based Method for Detecting Conveyor Belt Defects

by Qi Yang, Fang Li, Hong Tian, Hua Li, Shuai Xu, Jiyou Fei, Zhongkai Wu, Qiang Feng and Chang Lu

Appl. Sci. 2022, 12(19), 10051; https://doi.org/10.3390/app121910051 - 06 Oct 2022

Cited by 5 | Viewed by 1822

Abstract

Aiming to assess the problems of low detection accuracy, poor reliability, and high cost of the manual inspection method for conveyor-belt-surface defect detection, in this paper we propose a new method of conveyor-belt-surface defect detection based on knowledge distillation. First, a data enhancement [...] Read more.

Aiming to assess the problems of low detection accuracy, poor reliability, and high cost of the manual inspection method for conveyor-belt-surface defect detection, in this paper we propose a new method of conveyor-belt-surface defect detection based on knowledge distillation. First, a data enhancement method combining GAN and copy–pasting strategies is proposed to expand the dataset to solve the problem of insufficient and difficult-to-obtain samples of conveyor-belt-surface defects. Then, the target detection network, the YOLOv5 model, is pruned to generate a mini-network. A knowledge distillation method for fine-grained feature simulation is used to distill the lightweight detection network YOLOv5n and the pruned mini-network YOLOv5n-slim. The experiments show that our method significantly reduced the number of parameters and the inference time of the model, and significantly improves the detection accuracy, up to 97.33% accuracy, in the detection of conveyor belt defects. Full article

(This article belongs to the Special Issue Advanced Pattern Recognition & Computer Vision)

► Show Figures

Figure 1

Review

Jump to: Research

24 pages, 26974 KiB

Open AccessReview

A Review of Target Recognition Technology for Fruit Picking Robots: From Digital Image Processing to Deep Learning

by Xuehui Hua, Haoxin Li, Jinbin Zeng, Chongyang Han, Tianci Chen, Luxin Tang and Yuanqiang Luo

Appl. Sci. 2023, 13(7), 4160; https://doi.org/10.3390/app13074160 - 24 Mar 2023

Cited by 12 | Viewed by 3407

Abstract

Machine vision technology has dramatically improved the efficiency, speed, and quality of fruit-picking robots in complex environments. Target recognition technology for fruit is an integral part of the recognition systems of picking robots. The traditional digital image processing technology is a recognition method [...] Read more.

Machine vision technology has dramatically improved the efficiency, speed, and quality of fruit-picking robots in complex environments. Target recognition technology for fruit is an integral part of the recognition systems of picking robots. The traditional digital image processing technology is a recognition method based on hand-designed features, which makes it difficult to achieve better recognition as it results in dealing with the complex and changing orchard environment. Numerous pieces of literature have shown that extracting special features by training data with deep learning has significant advantages for fruit recognition in complex environments. In addition, to realize fully automated picking, reconstructing fruits in three dimensions is a necessary measure. In this paper, we systematically summarize the research work on target recognition techniques for picking robots in recent years, analyze the technical characteristics of different approaches, and conclude their development history. Finally, the challenges and future development trends of target recognition technology for picking robots are pointed out. Full article

(This article belongs to the Special Issue Advanced Pattern Recognition & Computer Vision)

► Show Figures

Figure 1

26 pages, 489 KiB

Open AccessReview

A Survey: Network Feature Measurement Based on Machine Learning

by Muyi Sun, Bingyu He, Ran Li, Jinhua Li and Xinchang Zhang

Appl. Sci. 2023, 13(4), 2551; https://doi.org/10.3390/app13042551 - 16 Feb 2023

Cited by 1 | Viewed by 2196

Abstract

In network management, network measuring is crucial. Accurate network measurements can increase network utilization, network management, and the ability to find network problems promptly. With extensive technological advancements, the difficulty for network measurement is not just the growth in users and traffic but [...] Read more.

In network management, network measuring is crucial. Accurate network measurements can increase network utilization, network management, and the ability to find network problems promptly. With extensive technological advancements, the difficulty for network measurement is not just the growth in users and traffic but also the increasingly difficult technical problems brought on by the network’s design becoming more complicated. In recent years, network feature measurement issues have been extensively solved by the use of ML approaches, which are ideally suited to thorough data analysis and the investigation of complicated network behavior. However, there is yet no favored learning model that can best address the network measurement issue. The problems that ML applications in the field of network measurement must overcome are discussed in this study, along with an analysis of the current characteristics of ML algorithms in network measurement. Finally, network measurement techniques that have been used as ML techniques are examined, and potential advancements in the field are explored and examined. Full article

(This article belongs to the Special Issue Advanced Pattern Recognition & Computer Vision)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Advanced Pattern Recognition & Computer Vision

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Published Papers (13 papers)

Research

Review

Further Information

Guidelines

MDPI Initiatives

Follow MDPI