Design of Gait Recognition System Based on FPGA and DSP

2014 ◽  
Vol 687-691 ◽  
pp. 3861-3868
Author(s):  
Zheng Hong Deng ◽  
Li Tao Jiao ◽  
Li Yan Liu ◽  
Shan Shan Zhao
Keyword(s):  
Monitoring System ◽  
Nearest Neighbor ◽  
Gait Recognition ◽  
Recognition Rate ◽  
Digital Signal ◽  
Recognition System ◽  
Recognition Algorithm ◽  
Zernike Moment ◽  
Intelligent Monitoring ◽  
The One

According to the trend of the intelligent monitoring system, on the basis of the study of gait recognition algorithm, the intelligent monitoring system is designed based on FPGA and DSP; On the one hand, FPGA’s flexibility and fast parallel processing algorithms when designing can be both used to avoid that circuit can not be modified after designed; On the other hand, the advantage of processing the digital signal of DSP is fully taken. In the feature extraction and recognition, Zernike moment is selected, at the same time the system uses the nearest neighbor classification method which is more mature and has good real-time performance. Experiments show that the system has high recognition rate.

Recognizing Check Magnetic Code Based on Peak-Valley Code and Distance

2011 ◽  
Vol 145 ◽  
pp. 588-592
Author(s):  
Di Fan ◽  
Jia Li ◽  
Mao Yong Cao ◽  
Nong Liang Sun ◽  
Hong Yu
Keyword(s):  
Character Recognition ◽  
Nearest Neighbor ◽  
Recognition Rate ◽  
Computational Cost ◽  
Recognition System ◽  
Recognition Algorithm ◽  
Statistical Parameters ◽  
Neighbor Recognition ◽  
Cash Payment ◽  
Low Computational Cost

The check is a popular form for the non-cash payment and accounts for more than 50% of the non-cash transactions. Magnetic ink character recognition system can recognize the check magnetic code automatically and get the information of the bank and account. In magnetic ink character recognition system, the recognizing algorithm is mostly based on correlation coefficient. The computational cost of this algorithm is very high. This paper has proposed a new algorithm based on the peak-valley code and peak-valley distance after analyzing the characteristics of magnetic code signals in E-13B standards to simplify the calculations and system design. Firstly, the magnetic code signal is normalized and separated into magnetic character signals by the thresholds of peak and valley. Secondly, the features of the peak-valley code and peak-valley distance of each magnetic character signal are extracted, then the recognition based on peak-valley code and the nearest neighbor recognition algorithm based on peak-valley distance are utilized to recognize the magnetic code. The recognition results and statistical parameters from a large number of experiments show that the new method has high recognition rate, good robustness and low computational cost.

Robustness for Authentication of the Human Using Face, Ear, and Gait Multimodal Biometric System

2021 ◽  
Vol 12 (1) ◽  
pp. 39-72
Author(s):  
Law Kumar Singh ◽  
Munish Khanna ◽  
Shankar Thawkar ◽  
Jagadeesh Gopal
Keyword(s):  
Feature Fusion ◽  
Gait Recognition ◽  
Recognition Rate ◽  
Recognition System ◽  
The Novel ◽  
Z Score ◽  
Biometric System ◽  
Multimodal Biometric System ◽  
Normalization Methods ◽  
Signature Recognition

Biometrics is the science that deals with personal human physiological and behavioral characteristics such as fingerprints, handprints, iris, voice, face recognition, signature recognition, ear recognition, and gait recognition. Recognition using a single trait has several problems and multimodal biometrics system is one of the solutions. In this work, the novel and imperative biometric feature gait is fused with face and ear biometric features for authentication and to overcome problems of the unimodal biometric recognition system. The authors have also applied various normalization methods to sort out the best solution for such a challenge. The feature fusion of the proposed multimodal biometric system has been tested using Min-Max and Z-score techniques. The computed results demonstrate that Z-Score outperforms the Min-Max technique. It is deduced that the Z-score is a promising method that generates a high recognition rate of 95% and a false acceptance rate of 10%.

A Reliability Design Methodology for Chinese Character Recognition

1998 ◽  
Vol 12 (02) ◽  
pp. 159-172 ◽  
Author(s):  
Y. S. Huang ◽  
K. Liu ◽  
C. Y. Suen ◽  
Y. Y. Tang
Keyword(s):  
Character Recognition ◽  
Chinese Character ◽  
Nearest Neighbor ◽  
Recognition Performance ◽  
Recognition Rate ◽  
Recognition System ◽  
Reliability Design ◽  
Chinese Character Recognition ◽  
Nearest Neighbor Rule ◽  
Pattern Distribution

This paper proposes a novel method which enables a Chinese character recognition system to obtain reliable recognition. In this method, two thresholds, i.e. class region thresholdRk and disambiguity thresholdAk, are used by each Chinese character k when the classifier is designed based on the nearest neighbor rule, where Rk defines the pattern distribution region of character k, and Ak prevents the samples not belonging to character k from being ambiguously recognized as character k. A novel algorithm to derive the appropriate thresholds Ak and Rk is developed so that a better recognition reliability can be obtained through iterative learning. Experiments performed on the ITRI printed Chinese character database have achieved highly reliable recognition performance (such as 0.999 reliability with a 95.14% recognition rate), which shows the feasibility and effectiveness of the proposed method.

Handwritten Character Recognition Based on a Multiple Fermat's Spiral

2013 ◽  
Vol 774-776 ◽  
pp. 1629-1635
Author(s):  
Aissa Boudjella ◽  
Brahim Belhouari Samir ◽  
Omar Kassem Khalil
Keyword(s):  
Character Recognition ◽  
High Performance ◽  
Nearest Neighbor ◽  
Recognition Rate ◽  
Recognition System ◽  
K Nearest Neighbor ◽  
Feature Extraction Method ◽  
Knn Classifier ◽  
New Feature ◽  
Performance Character

This paper describes a new feature extraction method which can be used very effectively in combination with Cluster K-Nearest Neighbor (CKNN) and KNN Classifier for image recognition. We propose handwritten English character recognition using Fermat's spiral approach to convert an image space into a parameter space. The system is implemented and simulated in MATLAB, and its performance is tested on real alphabet handwriting image. Fifteen (15) alphabet classes were created to carry out the experiment. Each class contains 9 alphabets {a, b, c, d, e, f, g, h, i}. The total of 15x9=135 alphabet images are captured under fixed camera position and controlled energy light intensity. The experimental results give a better recognition rate, 76.19% for KNN and 95.16% for C-KNN with reducing the overall data size of the transformed image. The relationship between the accuracy and k is investigated. It seems that when k goes from 1 to 9, the accuracy decreases linearly. The result of this investigation is a high performance character recognition system with significantly improved recognition rates and real-time.

Study of Intelligent Monitoring System of Low-Voltage Power Distribution

2013 ◽  
Vol 416-417 ◽  
pp. 781-784
Author(s):  
Mei Sun
Keyword(s):  
Monitoring System ◽  
Digital Signal Processor ◽  
Power Distribution ◽  
Distribution System ◽  
Low Voltage ◽  
Digital Signal ◽  
General Situation ◽  
Intelligent Monitoring ◽  
Power Distribution System ◽  
Calculation Ability

With the development of domestic low-voltage power distribution technology, people have an increasingly higher demand on the intelligence of low-voltage power distribution cabinet. Combined with the authors several years of experience of practice, this thesis first of all makes a brief analysis of the general situation of power distribution system automation, followed by a key analysis and conclusion of the characteristics of the existing low-voltage monitoring mode. Based on it, digital signal processor with strong floating point calculation ability, a new low-voltage intelligent monitoring system is designed.

scholarly journals Principal component analysis for human gait recognition system

2019 ◽  
Vol 8 (2) ◽  
pp. 569-576
Author(s):  
Othman O. Khalifa ◽  
Bilal Jawed ◽  
Sharif Shah Newaj Bhuiyn
Keyword(s):  
Principal Component Analysis ◽  
Minimum Distance ◽  
Gait Recognition ◽  
Principal Component ◽  
Initial Step ◽  
Component Analysis ◽  
Recognition System ◽  
Human Gait ◽  
The One ◽  
Main Components

This paper represents a method for Human Recognition system using Principal Component Analysis. Human Gait recognition works on the gait of walking subjects to identify people without them knowing or without their permission. The initial step in this kind of system is to generate silhouette frames of walking human. A number of features couldb be exytacted from these frames such as centriod ratio, heifht, width and orientation. The Principal Component Analysis (PCA) is used for the extracted features to condense the information and produces the main components that can represent the gait sequences for each waiking human. In the testing phase, the generated gait sequences are recognized by using a minimum distance classifier based on eluclidean distance matched with the one that already exist in the database used to identify walking subject.

scholarly journals Research and Implementation of Fast-LPRNet Algorithm for License Plate Recognition

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Zhichao Wang ◽  
Yu Jiang ◽  
Jiaxin Liu ◽  
Siyu Gong ◽  
Jian Yao ◽  
...  
Keyword(s):  
Traffic Management ◽  
Recognition Rate ◽  
Recognition System ◽  
Recognition Algorithm ◽  
License Plate ◽  
License Plate Recognition ◽  
Generalization Ability ◽  
Traffic Management System ◽  
High Recognition Accuracy ◽  
Fully Connected

The license plate recognition is an important part of the intelligent traffic management system, and the application of deep learning to the license plate recognition system can effectively improve the speed and accuracy of recognition. Aiming at the problems of traditional license plate recognition algorithms such as the low accuracy, slow speed, and the recognition rate being easily affected by the environment, a Convolutional Neural Network- (CNN-) based license plate recognition algorithm-Fast-LPRNet is proposed. This algorithm uses the nonsegment recognition method, removes the fully connected layer, and reduces the number of parameters. The algorithm—which has strong generalization ability, scalability, and robustness—performs license plate recognition on the FPGA hardware. Increaseing the depth of network on the basis of the Fast-LPRNet structure, the dataset of Chinese City Parking Dataset (CCPD) can be recognized with an accuracy beyond 90%. The experimental results show that the license plate recognition algorithm has high recognition accuracy, strong generalization ability, and good robustness.

Research on End-to-end Voiceprint Recognition Model Based on Convolutional Neural Network

2021 ◽  
Author(s):  
Hong Zhao ◽  
Lupeng Yue ◽  
Weijie Wang ◽  
Zeng Xiangyan
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Speech Signal ◽  
Recognition Rate ◽  
Recognition Algorithm ◽  
Recognition Model ◽  
One Dimensional ◽  
End To End ◽  
Original Speech ◽  
The One

Speech signal is a time-varying signal, which is greatly affected by individual and environment. In order to improve the end-to-end voice print recognition rate, it is necessary to preprocess the original speech signal to some extent. An end-to-end voiceprint recognition algorithm based on convolutional neural network is proposed. In this algorithm, the convolution and down-sampling of convolutional neural network are used to preprocess the speech signals in end-to-end voiceprint recognition. The one-dimensional and two-dimensional convolution operations were established to extract the characteristic parameters of Meier frequency cepstrum coefficient from the preprocessed signals, and the classical universal background model was used to model the recognition model of voice print. In this study, the principle of end-to-end voiceprint recognition was firstly analyzed, and the process of end-to-end voice print recognition, end-to-end voice print recognition features and Res-FD-CNN network structure were studied. Then the convolutional neural network recognition model was constructed, and the data were preprocessed to form the convolutional layer in frequency domain and the algorithm was tested.

WAVELET-BASED PALM VEIN RECOGNITION SYSTEM

2010 ◽  
Vol 03 (02) ◽  
pp. 131-134
Author(s):  
QIANG LI ◽  
YAN'AN ZENG ◽  
KUNTAO YANG
Keyword(s):  
Nearest Neighbor ◽  
Recognition Rate ◽  
Texture Feature ◽  
Recognition System ◽  
Vein Pattern ◽  
Personal Recognition ◽  
Vein Recognition ◽  
Palm Vein ◽  
Correct Recognition Rate ◽  
First Time

A new personal recognition system using the palm vein pattern is presented in this article. It is the first time that the palm vein pattern is used for personal recognition. The texture feature of palm vein is extracted by wavelet decomposition. With our palm vein image database, we employed the nearest neighbor (NN) classifier to test the performance of the system. Experimental results show that the algorithm based on wavelet transform can reach a correct recognition rate (CRR) of 98.8%.

scholarly journals A Novel Digital Modulation Recognition Algorithm Based on Deep Convolutional Neural Network

2020 ◽  
Vol 10 (3) ◽  
pp. 1166 ◽  
Author(s):  
Kaiyuan Jiang ◽  
Jiawei Zhang ◽  
Haibin Wu ◽  
Aili Wang ◽  
Yuji Iwahori
Keyword(s):  
Phase Shift ◽  
Rapid Development ◽  
Recognition Rate ◽  
Digital Signal ◽  
Recognition Algorithm ◽  
Svm Classifier ◽  
Phase Shift Keying ◽  
Modulation Recognition ◽  
Signal Modulation ◽  
Shift Keying

The modulation recognition of digital signals under non-cooperative conditions is one of the important research contents here. With the rapid development of artificial intelligence technology, deep learning theory is also increasingly being applied to the field of modulation recognition. In this paper, a novel digital signal modulation recognition algorithm is proposed, which has combined the InceptionResNetV2 network with transfer adaptation, called InceptionResnetV2-TA. Firstly, the received signal is preprocessed and generated the constellation diagram. Then, the constellation diagram is used as the input of the InceptionResNetV2 network to identify different kinds of signals. Transfer adaptation is used for feature extraction and SVM classifier is used to identify the modulation mode of digital signal. The constellation diagram of three typical signals, including Binary Phase Shift Keying(BPSK), Quadrature Phase Shift Keying(QPSK) and 8 Phase Shift Keying(8PSK), was made for the experiments. When the signal-to-noise ratio(SNR) is 4dB, the recognition rates of BPSK, QPSK and 8PSK are respectively 1.0, 0.9966 and 0.9633 obtained by InceptionResnetV2-TA, and at the same time, the recognition rate can be 3% higher than other algorithms. Compared with the traditional modulation recognition algorithms, the experimental results show that the proposed algorithm in this paper has a higher accuracy rate for digital signal modulation recognition at low SNR.

Sign in / Sign up

Export Citation Format

Share Document