scholarly journals Fast identification method for express end sorting label code based on convolutional recurrent neural network

2020 ◽  
Vol 14 (8) ◽  
pp. 1689-1697
Author(s):  
Haiyan Du ◽  
Chunxue Wu ◽  
Yan Wu ◽  
Ren Han ◽  
Xiao Lin ◽  
...  
Keyword(s):  
Neural Network ◽  
Recurrent Neural Network ◽  
Processing Speed ◽  
Recognition Accuracy ◽  
Code Region ◽  
Digital Information ◽  
Coding Region ◽  
Recognition Method ◽  
Sequence Recognition ◽  
Sorting Process

Abstract In the automatic sorting process of express, the express end sorting label code is used to indicate that the express is dispatched to a specific address by a specific courier. Since there are many areas on the express bill containing digital information, some areas may be improperly photographed, etc. The difficulty in positioning and recognizing the express end sorting label code region is increased. To solve this problem, this paper proposes an express end sorting label code recognition method with convolutional recurrent neural network for the code specification, which has certain versatility. In order to improve the overall code recognition speed, this paper optimizes the traditional digital recognition method, removes the original segmentation operation of the character and recognizes the code as sequence recognition. Firstly, the coding region is located, and then, the express end sorting label code is recognized by the convolutional recurrent neural network. In order to test the experimental performance, this paper tests on Free-Type dataset and SUN-synthesized dataset. The experimental results show that the proposed method improves the recognition accuracy and processing speed of the express end sorting label code.

scholarly journals Human Body Mixed Motion Pattern Recognition Method Based on Multi-Source Feature Parameter Fusion

Sensors ◽  
2020 ◽  
Vol 20 (2) ◽  
pp. 537 ◽  
Author(s):  
Jiyuan Song ◽  
Aibin Zhu ◽  
Yao Tu ◽  
Yingxu Wang ◽  
Muhammad Affan Arif ◽  
...  
Keyword(s):  
Neural Network ◽  
Pattern Recognition ◽  
Lower Extremity ◽  
Human Body ◽  
Recognition Accuracy ◽  
Screening Method ◽  
Motion Pattern ◽  
Recognition Method ◽  
Motion State ◽  
Motion Mode

Aiming at the requirement of rapid recognition of the wearer’s gait stage in the process of intelligent hybrid control of an exoskeleton, this paper studies the human body mixed motion pattern recognition technology based on multi-source feature parameters. We obtain information on human lower extremity acceleration and plantar analyze the relationship between these parameters and gait cycle studying the motion state recognition method based on feature evaluation and neural network. Based on the actual requirements of exoskeleton per use, 15 common gait patterns were determined. Using this, the studies were carried out on the time domain, frequency domain, and energy feature extraction of multi-source lower extremity motion information. The distance-based feature screening method was used to extract the optimal features. Finally, based on the multi-layer BP (back propagation) neural network, a nonlinear mapping model between feature quantity and motion state was established. The experimental results showed that the recognition accuracy in single motion mode can reach up to 98.28%, while the recognition accuracy of the two groups of experiments in mixed motion mode was found to be 92.7% and 97.4%, respectively. The feasibility and effectiveness of the model were verified.

scholarly journals A New Spiking Convolutional Recurrent Neural Network (SCRNN) With Applications to Event-Based Hand Gesture Recognition

2020 ◽  
Vol 14 ◽  
Author(s):  
Yannan Xing ◽  
Gaetano Di Caterina ◽  
John Soraghan
Keyword(s):  
Neural Network ◽  
Gesture Recognition ◽  
Recurrent Neural Network ◽  
Recognition Accuracy ◽  
Sequence Data ◽  
Arbitrary Length ◽  
Temporal Correlations ◽  
High Efficient ◽  
Recurrent Architecture ◽  
Event Based

The combination of neuromorphic visual sensors and spiking neural network offers a high efficient bio-inspired solution to real-world applications. However, processing event- based sequences remains challenging because of the nature of their asynchronism and sparsity behavior. In this paper, a novel spiking convolutional recurrent neural network (SCRNN) architecture that takes advantage of both convolution operation and recurrent connectivity to maintain the spatial and temporal relations from event-based sequence data are presented. The use of recurrent architecture enables the network to have a sampling window with an arbitrary length, allowing the network to exploit temporal correlations between event collections. Rather than standard ANN to SNN conversion techniques, the network utilizes a supervised Spike Layer Error Reassignment (SLAYER) training mechanism that allows the network to adapt to neuromorphic (event-based) data directly. The network structure is validated on the DVS gesture dataset and achieves a 10 class gesture recognition accuracy of 96.59% and an 11 class gesture recognition accuracy of 90.28%.

Drop Fingerprint Recognition Method Based on Cluster Analysis and BP Neural Network

2014 ◽  
Vol 543-547 ◽  
pp. 2099-2102 ◽  
Author(s):  
Qing Song ◽  
Dan Qing Du ◽  
Lu Yang ◽  
Gao Jie Meng ◽  
Xue Fei Mao
Keyword(s):  
Neural Network ◽  
Cluster Analysis ◽  
Bp Neural Network ◽  
Liquid Drop ◽  
Recognition Accuracy ◽  
Fingerprint Recognition ◽  
High Similarity ◽  
Recognition Method ◽  
Accuracy Rate ◽  
Accuracy Rates

The eigenvalues of some liquid drop fingerprints are of high similarity, which decreases the recognition accuracy rates of BP neural network. In order to solve this problem, recognition method based on cluster analysis and BP neural network is proposed in this paper. Cluster analysis is used to classify liquid samples according to the similarity of eigenvalues and narrow the recognition range for samples under study. The experimental results have proved that this method is able to increase the recognition accuracy rate from 83.42% to 99.83%.

scholarly journals Low Latency Based Convolutional Recurrent Neural Network Model for Speech Command Recognition

2021 ◽  
Vol 50 (4) ◽  
pp. 656-673
Author(s):  
Chhayarani Ram Kinkar ◽  
Yogendra Kumar Jain
Keyword(s):  
Neural Network ◽  
Recurrent Neural Network ◽  
Neural Network Model ◽  
Resource Constraints ◽  
Recognition Accuracy ◽  
Limited Resources ◽  
Low Latency ◽  
Model Parameters ◽  
Memory Requirement ◽  
Proposed Model

The presented paper proposes a new speech command recognition model for novel engineering applications with limited resources. We built the proposed model with the help of a Convolutional Recurrent Neural Network (CRNN). The use of CRNN instead of Convolutional Neural Network (CNN) helps us to reduce the model parameters and memory requirement as per resource constraints. Furthermore, we insert transmute and curtailment layer between the layers of CRNN. By doing this we further reduce model parameters and float number of operations to half of the CRNN requirement. The proposed model is tested on Google’s speech command dataset. The obtained result shows that the proposed CRNN model requires 1/3 parameters as compared to the CNN model. The number of parameters of the CRNN model is further reduced by 45% and the float numbers of operations between 2% to 12 % in different recognition tasks. The recognition accuracy of the proposed model is 96% on Google’s speech command dataset, and on laboratory recording, its recognition accuracy is 89%.

Study on Recognition Method of Ferrographic Debris Based on Neural Network

2013 ◽  
Vol 774-776 ◽  
pp. 1508-1511
Author(s):  
Ya Bin Cao
Keyword(s):  
Neural Network ◽  
Neural Network Model ◽  
Digital Image ◽  
Bp Neural Network ◽  
Recognition Accuracy ◽  
Characteristic Parameter ◽  
Field Of View ◽  
Recognition Method ◽  
Lighting Conditions ◽  
Bp Neural Network Model

A new method is proposed in which three images of the same field of view under different lighting conditions were analysed comprehensively. On the basis of pretreatment of debris digital image and extraction of characteristic parameter of debrises, the debrises were recognized by establishing the BP neural network model. The results showed that the recognition accuracy of debrises could reach 85.6% with this method.

scholarly journals Cable Incipient Fault Identification with a Sparse Autoencoder and a Deep Belief Network

Energies ◽  
2019 ◽  
Vol 12 (18) ◽  
pp. 3424 ◽  
Author(s):  
Ning Liu ◽  
Bo Fan ◽  
Xianyong Xiao ◽  
Xiaomei Yang
Keyword(s):  
Neural Network ◽  
Pattern Recognition ◽  
Recognition Accuracy ◽  
Power Cables ◽  
Pattern Recognition Method ◽  
Recognition Method ◽  
Belief Network ◽  
Incipient Faults ◽  
Sparse Autoencoder ◽  
Traditional Pattern

Incipient faults in power cables are a serious threat to power safety and are difficult to accurately identify. The traditional pattern recognition method based on feature extraction and feature selection has strong subjectivity. If the key feature information cannot be extracted accurately, the recognition accuracy will directly decrease. To accurately identify incipient faults in power cables, this paper combines a sparse autoencoder and a deep belief network to form a deep neural network, which relies on the powerful learning ability of the neural network to classify and identify various cable fault signals, without requiring preprocessing operations for the fault signals. The experimental results demonstrate that the proposed approach can effectively identify cable incipient faults from other disturbances with a similar overcurrent phenomenon and has a higher recognition accuracy and reliability than the traditional pattern recognition method.

scholarly journals Development of the malware detection model based on an analysis of API request sequences

2021 ◽  
pp. 90-96
Author(s):  
Александр Николаевич Черкасов ◽  
Егор Алексеевич Туркин
Keyword(s):  
Neural Network ◽  
Network Model ◽  
Recurrent Neural Network ◽  
Neural Network Model ◽  
Complex Analysis ◽  
Recognition Accuracy ◽  
Network Algorithms ◽  
Malicious Software ◽  
Detection Model ◽  
Optimal Approach

Анализ актуальности существующих работ по изучению нейросетевых алгоритмов обнаружения вредоносного программного обеспечения позволил определить оптимальный подход к решению задачи распознавания вирусов. Был сформирован набор данных, содержащий последовательности системных запросов для 2083 исполняемых файлов. Обучение модели рекуррентной нейронной сети на полученном наборе данных позволило получить классификатор с точностью распознавания вредоносных программ 97,6%. Ключевой аспект значимости работы заключается в практической ориентированности исследования - полученная модель будет применена в качестве компонента системы комплексного анализа вредоносного программного обеспечения. An analysis of the relevance of existing works on the study of neural network algorithms for detecting malicious software made it possible to determine the optimal approach to solving the problem of virus recognition. A dataset has been generated containing sequences of system requests for 2083 executable files. Training the recurrent neural network model on the generated dataset made it possible to obtain a classifier with a malware recognition accuracy of 97.6%. The key aspect of the significance of the work lies in the practical orientation of the research - the resulting model will be applied as a component of a system for complex analysis of malicious software.

scholarly journals Human Recognition Using Joint Coordinate Images (JCIs) with Convolutional Neural Network

2021 ◽  
Vol 2095 (1) ◽  
pp. 012056
Author(s):  
Deyu Kong ◽  
Xuejun Zhang ◽  
Yini Wei ◽  
Xianfu Xu ◽  
Hongjie Zeng ◽  
...  
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Recognition Accuracy ◽  
Color Image ◽  
Recognition Task ◽  
Skeletal Structure ◽  
Neural Network Classifier ◽  
Recognition Method ◽  
Human Recognition ◽  
Complementary Method

Abstract Human recognition with skeletal data has the advantage in detecting people without their face characteristic on image. However, the accuracy of recognition by this method is always low because it relies deeply on manual feature selection. We propose a novel human recognition method called Joint Coordinate Images (JCIs) with Convolutional Neural Network (CNN) based on the image generated from skeletal information tracked by KinectV1. In order to represent human physical skeletal characteristic, the coordinate values XYZ of human joint tracked by KinectV1 are firstly created in color image called Joint Coordinate Images (JCIs), in which the relative position of the pixels represents the skeletal structure characteristics of participants with shape in “大” structure. Secondly, a new convolution neural network classifier Lenet-5 model, which always performed well in image classification, was modified to be able to input our JCIs for human recognition. The experimental results show that human recognition using joint coordinate image and Lenet-5 network can reach the highest recognition accuracy of 90.00% on the G3D dataset, which demonstrates the feasibility to transform the skeletal coordinate information into color image for human recognition task and could be used as a complementary method to the well-known application of face recognition.

Sign in / Sign up

Export Citation Format

Share Document