scholarly journals Underwater Target Recognition Based on Improved YOLOv4 Neural Network

Electronics ◽  
2021 ◽  
Vol 10 (14) ◽  
pp. 1634
Author(s):  
Lingyu Chen ◽  
Meicheng Zheng ◽  
Shunqiang Duan ◽  
Weilin Luo ◽  
Ligang Yao
Keyword(s):  
Neural Network ◽  
Target Recognition ◽  
Data Preprocessing ◽  
Objective Evaluation ◽  
The Other ◽  
Training Set ◽  
Underwater Target ◽  
Hardware Performance

The YOLOv4 neural network is employed for underwater target recognition. To improve the accuracy and speed of recognition, the structure of YOLOv4 is modified by replacing the upsampling module with a deconvolution module and by incorporating depthwise separable convolution into the network. Moreover, the training set used in the YOLO network is preprocessed by using a modified mosaic augmentation, in which the gray world algorithm is used to derive two images when performing mosaic augmentation. The recognition results and the comparison with the other target detectors demonstrate the effectiveness of the proposed YOLOv4 structure and the method of data preprocessing. According to both subjective and objective evaluation, the proposed target recognition strategy can effectively improve the accuracy and speed of underwater target recognition and reduce the requirement of hardware performance as well.

scholarly journals Sonar image recognition of underwater target based on convolutional neural network

2021 ◽  
Vol 39 (2) ◽  
pp. 285-291
Author(s):  
Leilei Jin ◽  
Hong LIANG ◽  
Changsheng Yang
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Image Recognition ◽  
Target Recognition ◽  
Saliency Detection ◽  
Image Features ◽  
Feature Maps ◽  
Sonar Image ◽  
Sonar Images ◽  
Underwater Target

Underwater target recognition is one core technology of underwater unmanned detection. To improve the accuracy of underwater automatic target recognition, a sonar image recognition method based on convolutional neural network was proposed and the underwater target recognition model was established according to the characteristics of sonar images. Firstly, the sonar image was segmented and clipped with a saliency detection method to reduce the dimension of input data, and to reduce the interference of image background to the feature extraction process. Secondly, by using stacked convolutional layers and pooling layers, the high-level semantic information of the target was automatically learned from the input sonar image, to avoid damaging the effective information caused by extracting image features manually. Finally, the spatial pyramid pooling method was used to extract the multi-scale information from the sonar feature maps, which was to make up for the lack of detailed information of sonar images and solve the problem caused by the inconsistent size of input images. On the collected sonar image dataset, the experimental results show that the target recognition accuracy of the present method can recognize underwater targets more accurately and efficiently than the conventional convolutional neural networks.

scholarly journals Underwater target recognition method based on convolution residual network

2019 ◽  
Vol 283 ◽  
pp. 04011
Author(s):  
Yuechao Chen ◽  
Shuanping Du ◽  
HengHeng Quan ◽  
Bin Zhou
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Recognition Accuracy ◽  
Deep Structure ◽  
Target Recognition ◽  
Complex Signal ◽  
Recognition Method ◽  
Residual Network ◽  
Time Frequency ◽  
Underwater Target

The underwater target radiated noises usually have characteristics of low signal to noise ratio, complex signal components and so on. Therefore the recognition is a difficult task and powerful recognition method must be applied to obtain good results. In this paper, a recognition method for underwater target radiated noise time-frequency image based on convolutional neural network with residual units is proposed. The principles and characteristics of the convolutional residual network are analyzed and three basic convolutional residual units are put forward. Then three convolutional residual network models with very deep structure are established based on basic convolutional residual units and some normal convolution layers. The number of the hidden layers is 50, 100 and 150 respectively and softmax algorithm is used as the top classifier. The wavelet transform is adopted to generate time-frequency images of the underwater target radiated noises with frequency band of 10~200Hz, thus ensuring the accuracy of local structure of the image, then the above three models can be used to recognize the images. The experimental data of two types of targets were processed. The results are as follows. As the number of training time increases, the training loss shows a convergence trend and the recognition accuracy of test data gradually increases to more than 90%. In addition, the top-level output has obvious separability. The final recognition accuracies of the three convolutional residual networks are all over 93% and higher than that of normal convolutional neural network with 5 layers. As the number of layers increases, the recognition accuracy of the convolutional residual network increases to a certain extent, illustrating the increase of layer number can improve the processing effect. The analysis results show that the convolution residual network can extract features with separability through deep structure and achieve effective underwater target recognition.

scholarly journals ia-PNCC: Noise Processing Method for Underwater Target Recognition Convolutional Neural Network

2019 ◽  
Vol 58 (1) ◽  
pp. 169-181 ◽  
Author(s):  
Nianbin Wang ◽  
Ming He ◽  
Jianguo Sun ◽  
Hongbin Wang ◽  
Lianke Zhou ◽  
...  
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Target Recognition ◽  
Processing Method ◽  
Underwater Target

scholarly journals Geometric property-based convolutional neural network for indoor object detection

2021 ◽  
Vol 18 (1) ◽  
pp. 172988142199332
Author(s):  
Xintao Ding ◽  
Boquan Li ◽  
Jinbao Wang
Keyword(s):  
Neural Network ◽  
Object Detection ◽  
Convolutional Neural Network ◽  
Geometric Property ◽  
Ground Truth ◽  
Geometric Constraints ◽  
Depth Information ◽  
Training Set ◽  
Object Knowledge ◽  
The Mean

Indoor object detection is a very demanding and important task for robot applications. Object knowledge, such as two-dimensional (2D) shape and depth information, may be helpful for detection. In this article, we focus on region-based convolutional neural network (CNN) detector and propose a geometric property-based Faster R-CNN method (GP-Faster) for indoor object detection. GP-Faster incorporates geometric property in Faster R-CNN to improve the detection performance. In detail, we first use mesh grids that are the intersections of direct and inverse proportion functions to generate appropriate anchors for indoor objects. After the anchors are regressed to the regions of interest produced by a region proposal network (RPN-RoIs), we then use 2D geometric constraints to refine the RPN-RoIs, in which the 2D constraint of every classification is a convex hull region enclosing the width and height coordinates of the ground-truth boxes on the training set. Comparison experiments are implemented on two indoor datasets SUN2012 and NYUv2. Since the depth information is available in NYUv2, we involve depth constraints in GP-Faster and propose 3D geometric property-based Faster R-CNN (DGP-Faster) on NYUv2. The experimental results show that both GP-Faster and DGP-Faster increase the performance of the mean average precision.

scholarly journals Fast Accurate and Automatic Brushstroke Extraction

2021 ◽  
Vol 17 (2) ◽  
pp. 1-24
Author(s):  
Yunfei Fu ◽  
Hongchuan Yu ◽  
Chih-Kuo Yeh ◽  
Tong-Yee Lee ◽  
Jian J. Zhang
Keyword(s):  
Neural Network ◽  
Efficient Algorithm ◽  
Deep Neural Network ◽  
High Efficiency ◽  
State Of The Art ◽  
High Reliability ◽  
The Other ◽  
Manual Annotation ◽  
Stroke Extraction ◽  
Art Research

Brushstrokes are viewed as the artist’s “handwriting” in a painting. In many applications such as style learning and transfer, mimicking painting, and painting authentication, it is highly desired to quantitatively and accurately identify brushstroke characteristics from old masters’ pieces using computer programs. However, due to the nature of hundreds or thousands of intermingling brushstrokes in the painting, it still remains challenging. This article proposes an efficient algorithm for brush Stroke extraction based on a Deep neural network, i.e., DStroke. Compared to the state-of-the-art research, the main merit of the proposed DStroke is to automatically and rapidly extract brushstrokes from a painting without manual annotation, while accurately approximating the real brushstrokes with high reliability. Herein, recovering the faithful soft transitions between brushstrokes is often ignored by the other methods. In fact, the details of brushstrokes in a master piece of painting (e.g., shapes, colors, texture, overlaps) are highly desired by artists since they hold promise to enhance and extend the artists’ powers, just like microscopes extend biologists’ powers. To demonstrate the high efficiency of the proposed DStroke, we perform it on a set of real scans of paintings and a set of synthetic paintings, respectively. Experiments show that the proposed DStroke is noticeably faster and more accurate at identifying and extracting brushstrokes, outperforming the other methods.

Sign in / Sign up

Export Citation Format

Share Document