scholarly journals E-Commerce Picture Text Recognition Information System Based on Deep Learning

2022 ◽  
Vol 2022 ◽  
pp. 1-11
Author(s):  
Bin Zhao ◽  
WenYing Li ◽  
Qian Guo ◽  
RongRong Song
Keyword(s):  
Neural Network ◽  
Target Detection ◽  
Recognition Accuracy ◽  
Text Recognition ◽  
Detection Accuracy ◽  
Image Detection ◽  
Evaluation Indexes ◽  
Different Levels ◽  
Made In ◽  
Detection And Recognition

For the accuracy requirements of commodity image detection and classification, the FPN network is improved by DPFM ablation and RFM, so as to improve the detection accuracy of commodities by the network. At the same time, in view of the narrowing of channels in the application of traditional MWI-DenseNet network, a new GTNet network is proposed to improve the classification accuracy of commodities.The results show that at different levels of evaluation indexes, the dpFPN-Netv2 algorithm improved by DPFM + RFM fusion has higher target detection accuracy than RetinaNet-50 algorithm and other algorithms. And the detection time is 52 ms, which is significantly lower than 90 ms required for RetinaNet-50 detection. In terms of target recognition, compared with the traditional MWI-DenseNet neural network, the computation amount of the improved MWI DenseNet neural network is significantly reduced under different shunt ratios, and the recognition accuracy is significantly improved. The innovation of this study lies in improving the algorithm from the perspective of target detection and recognition, so as to change the previous improvement that only can be made in a single way.

scholarly journals Image target detection algorithm compression and pruning based on neural network

2021 ◽  
Vol 18 (2) ◽  
pp. 499-516
Author(s):  
Yan Sun ◽  
Zheping Yan
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Target Detection ◽  
Network Model ◽  
Detection Algorithm ◽  
Image Feature ◽  
Propagation Mode ◽  
Learning Technology ◽  
Image Detection ◽  
Detection Algorithms

The main purpose of target detection is to identify and locate targets from still images or video sequences. It is one of the key tasks in the field of computer vision. With the continuous breakthrough of deep machine learning technology, especially the convolutional neural network model shows strong Ability to extract image feature in the field of digital image processing. Although the model research of target detection based on convolutional neural network is developing rapidly, but there are still some problems in practical applications. For example, a large number of parameters requires high storage and computational costs in detected model. Therefore, this paper optimizes and compresses some algorithms by using early image detection algorithms and image detection algorithms based on convolutional neural networks. After training and learning, there will appear forward propagation mode in the application of CNN network model, providing the model for image feature extraction, integration processing and feature mapping. The use of back propagation makes the CNN network model have the ability to optimize learning and compressed algorithm. Then research discuss the Faster-RCNN algorithm and the YOLO algorithm. Aiming at the problem of the candidate frame is not significant which extracted in the Faster- RCNN algorithm, a target detection model based on the Significant area recommendation network is proposed. The weight of the feature map is calculated by the model, which enhances the saliency of the feature and reduces the background interference. Experiments show that the image detection algorithm based on compressed neural network image has certain feasibility.

scholarly journals A New Method of Different Neural Network Depth and Feature Map Size on Remote Sensing Small Target Detection

2021 ◽  
Vol 24 (68) ◽  
pp. 21-32
Author(s):  
Yaming Cao ◽  
ZHEN YANG ◽  
CHEN GAO
Keyword(s):  
Neural Network ◽  
Remote Sensing ◽  
Neural Networks ◽  
Target Detection ◽  
Detection Efficiency ◽  
Detection Accuracy ◽  
Small Target ◽  
Remote Sensing Images ◽  
Feature Maps ◽  
Small Target Detection

Convolutional neural networks (CNNs) have shown strong learning capabilities in computer vision tasks such as classification and detection. Especially with the introduction of excellent detection models such as YOLO (V1, V2 and V3) and Faster R-CNN, CNNs have greatly improved detection efficiency and accuracy. However, due to the special angle of view, small size, few features, and complicated background, CNNs that performs well in the ground perspective dataset, fails to reach a good detection accuracy in the remote sensing image dataset. To this end, based on the YOLO V3 model, we used feature maps of different depths as detection outputs to explore the reasons for the poor detection rate of small targets in remote sensing images by deep neural networks. We also analyzed the effect of neural network depth on small target detection, and found that the excessive deep semantic information of neural network has little effect on small target detection. Finally, the verification on the VEDAI dataset shows, that the fusion of shallow feature maps with precise location information and deep feature maps with rich semantics in the CNNs can effectively improve the accuracy of small target detection in remote sensing images.

scholarly journals Detection of ship targets in photoelectric images based on an improved recurrent attention convolutional neural network

2020 ◽  
Vol 16 (3) ◽  
pp. 155014772091295 ◽  
Author(s):  
Zhijing Xu ◽  
Yuhao Huo ◽  
Kun Liu ◽  
Sidong Liu
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Image Feature ◽  
Detection Accuracy ◽  
Image Detection ◽  
Image Feature Extraction ◽  
Ship Detection ◽  
Multi Scale ◽  
Network Methods ◽  
Accelerate Convergence

Deep learning algorithms have been increasingly used in ship image detection and classification. To improve the ship detection and classification in photoelectric images, an improved recurrent attention convolutional neural network is proposed. The proposed network has a multi-scale architecture and consists of three cascading sub-networks, each with a VGG19 network for image feature extraction and an attention proposal network for locating feature area. A scale-dependent pooling algorithm is designed to select an appropriate convolution in the VGG19 network for classification, and a multi-feature mechanism is introduced in attention proposal network to describe the feature regions. The VGG19 and attention proposal network are cross-trained to accelerate convergence and to improve detection accuracy. The proposed method is trained and validated on a self-built ship database and effectively improve the detection accuracy to 86.7% outperforming the baseline VGG19 and recurrent attention convolutional neural network methods.

scholarly journals An Efficient Deep Convolutional Neural Network Approach for Object Detection and Recognition Using a Multi-Scale Anchor Box in Real-Time

2021 ◽  
Vol 13 (12) ◽  
pp. 307
Author(s):  
Vijayakumar Varadarajan ◽  
Dweepna Garg ◽  
Ketan Kotecha
Keyword(s):  
Neural Network ◽  
Object Detection ◽  
Real Time ◽  
Semantic Segmentation ◽  
Object Identification ◽  
Detection Accuracy ◽  
Neural Network Approach ◽  
Multi Scale ◽  
Scale Anchor ◽  
Detection And Recognition

Deep learning is a relatively new branch of machine learning in which computers are taught to recognize patterns in massive volumes of data. It primarily describes learning at various levels of representation, which aids in understanding data that includes text, voice, and visuals. Convolutional neural networks have been used to solve challenges in computer vision, including object identification, image classification, semantic segmentation and a lot more. Object detection in videos involves confirming the presence of the object in the image or video and then locating it accurately for recognition. In the video, modelling techniques suffer from high computation and memory costs, which may decrease performance measures such as accuracy and efficiency to identify the object accurately in real-time. The current object detection technique based on a deep convolution neural network requires executing multilevel convolution and pooling operations on the entire image to extract deep semantic properties from it. For large objects, detection models can provide superior results; however, those models fail to detect the varying size of the objects that have low resolution and are greatly influenced by noise because the features after the repeated convolution operations of existing models do not fully represent the essential characteristics of the objects in real-time. With the help of a multi-scale anchor box, the proposed approach reported in this paper enhances the detection accuracy by extracting features at multiple convolution levels of the object. The major contribution of this paper is to design a model to understand better the parameters and the hyper-parameters which affect the detection and the recognition of objects of varying sizes and shapes, and to achieve real-time object detection and recognition speeds by improving accuracy. The proposed model has achieved 84.49 mAP on the test set of the Pascal VOC-2007 dataset at 11 FPS, which is comparatively better than other real-time object detection models.

scholarly journals Research on Methods of English Text Detection and Recognition Based on Neural Network Detection Model

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Chunlan Li
Keyword(s):  
Neural Network ◽  
Recognition Accuracy ◽  
Rapid Development ◽  
Research Work ◽  
Text Detection ◽  
English Text ◽  
Data Sets ◽  
Original Algorithm ◽  
Multiple Data Sets ◽  
Detection And Recognition

With the rapid development of computer science, a large number of images and an explosive amount of information make it difficult to filter and effectively extract information. This article focuses on the inability of effective detection and recognition of English text content to conduct research, which is useful for improving the application of intelligent analysis significance. This paper studies how to improve the neural network model to improve the efficiency of image text detection and recognition under complex background. The main research work is as follows: (1) An improved CTPN multidirectional text detection algorithm is proposed, and the algorithm is applied to the multidirectional text detection and recognition system. It uses the multiangle rotation of the image to be detected, then fuses the candidate text boxes detected by the CTPN network, and uses the fusion strategy to find the best area of the text. This algorithm solves the problem that the CTPN network can only detect the text in the approximate horizontal direction. (2) An improved CRNN text recognition algorithm is proposed. The algorithm is based on CRNN and combines traditional text features and depth features at the same time, making it possible to recognize occluded text. The algorithm was tested on the IC13 and SVT data sets. Compared with the CRNN algorithm, the recognition accuracy has been improved, and the detection and recognition accuracy has increased by 0.065. This paper verifies the effectiveness of the improved algorithm model on multiple data sets, which can effectively detect various English texts, and greatly improves the detection and recognition performance of the original algorithm.

scholarly journals Improving Neural Network Detection Accuracy of Electric Power Bushings in Infrared Images by Hough Transform

Sensors ◽  
2020 ◽  
Vol 20 (10) ◽  
pp. 2931
Author(s):  
Hongshan Zhao ◽  
Zeyan Zhang
Keyword(s):  
Neural Network ◽  
Electric Power ◽  
Target Detection ◽  
Hough Transform ◽  
Image Method ◽  
Detection Accuracy ◽  
Infrared Images ◽  
Gap Statistic ◽  
The Neural Network ◽  
Bounding Boxes

To improve the neural network detection accuracy of the electric power bushings in infrared images, a modified algorithm based on the You Only Look Once version 2 (YOLOv2) network is proposed to achieve better recognition results. Specifically, YOLOv2 corresponds to a convolutional neural network (CNN), although its rotation invariance is poor, and some bounding boxes (BBs) exhibit certain deviations. To solve this problem, the standard Hough transform and image rotation are utilized to determine the optimal recognition angle for target detection, such that an optimal recognition effect of YOLOv2 on inclined objects (for example, bushing) is achieved. With respect to the problem that the BB is biased, the shape feature of the bushing is extracted by the Gap statistic algorithm, based on K-means clustering; thereafter, the sliding window (SW) is utilized to determine the optimal recognition area. Experimental verification indicates that the proposed rotating image method can improve the recognition effect, and the SW can further modify the BB. The accuracy of target detection increases to 97.33%, and the recall increases to 95%.

scholarly journals Ship Target Detection Algorithm Based on Improved Faster R-CNN

Electronics ◽  
2019 ◽  
Vol 8 (9) ◽  
pp. 959 ◽  
Author(s):  
Qi ◽  
Li ◽  
Chen ◽  
Wang ◽  
Dong ◽  
...  
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Target Detection ◽  
Texture Features ◽  
Detection Algorithm ◽  
Detection Accuracy ◽  
Marine Transportation ◽  
Computational Speed ◽  
Neural Network Algorithm ◽  
Detection Search

Ship target detection has urgent needs and broad application prospects in military and marine transportation. In order to improve the accuracy and efficiency of the ship target detection, an improved Faster R-CNN (Faster Region-based Convolutional Neural Network) algorithm of ship target detection is proposed. In the proposed method, the image downscaling method is used to enhance the useful information of the ship image. The scene narrowing technique is used to construct the target regional positioning network and the Faster R-CNN convolutional neural network into a hierarchical narrowing network, aiming at reducing the target detection search scale and improving the computational speed of Faster R-CNN. Furthermore, deep cooperation between main network and subnet is realized to optimize network parameters after researching Faster R-CNN with subject narrowing function and selecting texture features and spatial difference features as narrowed sub-networks. The experimental results show that the proposed method can significantly shorten the detection time of the algorithm while improving the detection accuracy of Faster R-CNN algorithm.

Sign in / Sign up

Export Citation Format

Share Document