Proposed Recognition System Research Articles

LPI Radar Waveform Recognition Based on Multi-Branch MWC Compressed Sampling Receiver

In this paper, a new wideband digital receiver based on the modulated wideband converter compressed sampling (CS) system is proposed to replace a conventional wideband digital receiver, solve the cross-channel signal problem, and achieve intra-pulse modulation recognition for the low probability of intercept (LPI) radar signals. The proposed receiver uses cyclic-shifted pseudo-random sequences to mix the received signals to baseband. The mixed signals are low-pass filtered and down-sampled to obtain compressed sampling data containing the full information of the received signals. Since the phases of the multi-branch CS data are designed to change regularly, we propose a phase correction factor to correct the phases of the multi-branch CS data, which can be superposed to increase the output signal-to-noise ratio (SNR). Then, a recognition method based on the short-time Fourier transform (STFT) and spectrum energy focusing rate tests are proposed. First, the spectrum modulation bandwidth based on the STFT spectrum of the superposed CS data is tested to distinguish phase modulation signals and frequency modulation signals approximately. Then, the spectrum energy focusing rate of the superposed CS data is tested to determine the intra-pulse modulation type specifically. Finally, the superiority of the proposed recognition system for LPI radar signals and cross-channel signals is demonstrated by simulations and analysis. Simulation results show that the overall ratio of the successful recognition (RSR) is 100% when the SNR is greater than 0 dB.

Probabilistic Model Based Hybrid Classifier for Character Recognition

Handwritten character recognition is most crucial one indulging in many of the applications like forensic search, searching historical manuscripts, mail sorting, bank check reading, tax form processing, book and handwritten notes transcription etc. The problem occurrence in the recognition is mainly because of the writing style variation, size variation (length and height), orientation angle etc. In this paper a probabilistic model based hybrid classifier is proposed for the character recognition combining the neural network and decision tree classifiers. In addition to the local gradient features i.e. histogram oriented feature and grid level feature, an additional feature called GLCM feature is extracted from the input image in the proposed recognition system and are concatenated for the image recognition procedure to encode color, shape, texture, local as well as the statistical information. These extracted features considered are given to the hybrid classifier which recognises the character. In the test set, recognition accuracy of 95% is achieved. The proposed probabilistic model based hybrid classifier tends to contribute more accurate character recognition rate compared to the existing character recognition system.

Iraqi Kurdistan Vehicle License Plate Recognition System based on Client-server Network

Vehicle license plate recognition (VLPR) system has a wide range of applications in real life, such as parking lots, private and public entrances, theft and border control. This paper presents a system overview of client-server license plate recognition system for vehicles in Kurdistan Region of Iraq besides, the vehicle type and city in each plate is identified. The client/server model consist of two components called client-side and server-side. The images are provided by the client side and will be sent over the server to be processed. The process of license plate recognition is done in the server side with the integration with the Matlab software. The proposed recognition system is developed based on digital images and the main component of the system consist of: preprocessing, license plate detection, segmentation and recognition process. A preprocessing step using image enhancement, filtering, and adaptive thresholding is employed. For localizing and extracting the license plate region, Haar wavelet transformation and edge detection algorithm are used. license plate number segmentation is carried out using bounding box based morphological operations and finally color histogram technique and correlation coefficient based similarity measurement are used for recognition task. The experimental results show that our proposed system for the license plate detection and recognition process have obviously higher recognition accuracy rate.

Motion intent recognition of individual fingers based on mechanomyogram

The mechanomyogram (MMG) signals detected from forearm muscle group contain abundant information which can be utilized to predict finger motion intention. Few works have been reported in this area especially for the recognition of individual finger motions, which however is crucial for many applications such as prosthesis control. In this paper, a MMG based finger gesture recognition system is designed to identify the motions of each finger. In this system, three kinds of feature sets, wavelet packet transform (WPT) coefficients, stationary wavelet transform (SWT) coefficients, and the time and frequency domain hybrid (TFDH) features, are adopted and processed by a support vector machine (SVM) classifier. The experimental results show that the average accuracy rates of recognition using the WPT, SWT and TFDH features are 91.64%, 94.31%, and 91.56%, respectively. Furthermore, the average rate of 95.20% can be achieved when above three feature sets are combined to use in the proposed recognition system.

Off-line Handwritten Numeral Recognition using Hybrid Feature Set – A Comparative Analysis

Handwritten numeral recognition has always been a very challenging task due to many variations in handwritten numerals with different writing styles. It is an active research area now a day. To tackle these variations and to get optimal recognition results, a hybrid feature set, which consists of multiple feature extraction approaches like Box Method, Mean, Standard Deviation and Centre of Gravity, has been used in this paper for recognizing the handwritten numerals. A Neural network has been used for successfully classifying 550 samples taken from “The Chars74” handwritten numerals dataset. The appropriate number of hidden neurons and different membership functions has been used to enhance the recognition results. The proposed recognition system is evaluated and compared with other methods.

Offline Signature Recognition and Verification System using Efficient Fuzzy Kohonen Clustering Network (EFKCN) Algorithm

Research on offline signature recognition still has not shown satisfactory results as the results of recent research. Therefore this study aims to proposed an offline signature recognition and verification system which employed an efficient fuzzy Kohonen clustering networks (EFKCN)1 algorithm. The proposed recognition system and signature verification system consist of five stages including data acquisition, image processing, data normalization, clustering, and evaluation. The recognition of signature patterns using the clustering method with the EFKCN algorithm shows relatively better result with 70% accuracy compared to the accuracy of previous research results2 which is 53%, and a good signature recognition result can be developed to assist the verification system as well as the personal data verification system as made in this study.

Unconstrained handwritten digit recognition using perceptual shape primitives

In this paper, we propose a new handwritten digit recognition method which works in a very similar way as human perception. The digit image boundary is decomposed into four salient visual primitives, namely closure, smooth curve, protrusion and straight segment by defining a set of external symmetry axis. Unlike the conventional algorithms, our low complexity shape decomposition method neither searches for curvature minima nor finds optimal parsing by using shortcut and convexity rules. Based on the spatial configuration of extracted primitives, the recognizer classifies a test digit image using a set of classification rules. The performance of our proposed recognition system is evaluated on five digit datasets of four popular scripts, Odia, Bangla, Arabic and English. The recognition accuracies on the ISI Kolkata Odia and Bangla, IITBBS Odia, CMATERdb Arabic and MNIST English digit datasets are found to be 99.02, 99.25, 99.66, 97.96 and 99.11%, respectively. The proposed method outperforms the existing recognition systems on both the Odia digit datasets and achieves comparable performance in other cases.

A Survey of 2D Face Recognition Techniques

Despite the existence of various biometric techniques, like fingerprints, iris scan, as well as hand geometry, the most efficient and more widely-used one is face recognition. This is because it is inexpensive, non-intrusive and natural. Therefore, researchers have developed dozens of face recognition techniques over the last few years. These techniques can generally be divided into three categories, based on the face data processing methodology. There are methods that use the entire face as input data for the proposed recognition system, methods that do not consider the whole face, but only some features or areas of the face and methods that use global and local face characteristics simultaneously. In this paper, we present an overview of some well-known methods in each of these categories. First, we expose the benefits of, as well as the challenges to the use of face recognition as a biometric tool. Then, we present a detailed survey of the well-known methods by expressing each method’s principle. After that, a comparison between the three categories of face recognition techniques is provided. Furthermore, the databases used in face recognition are mentioned, and some results of the applications of these methods on face recognition databases are presented. Finally, we highlight some new promising research directions that have recently appeared.

A robust and automatic recognition system of analog instruments in power system by using computer vision

Abstract It is a puzzling issue found in the field of Supervisory Control and Data Acquisition (SCADA) to read automatically the indication of analog measuring instruments, which collects many states of power system equipment. To address this problem, a lot of approaches based on computer vision have been proposed for the automatic reading of the analog measuring instruments indication. However, most the proposed systems are sensitive to brightness. Moreover, since the camera angle must be vertical, the robustness of the approaches cannot meet application requirements. In this paper, the proposed recognition system uses a novel automatic reading approach, which can automatically read the indication of analog measuring instruments at various brightness levels and camera angles. It is an integrated application of Multi-Scale Retinex with color restoration at different brightness levels, Perspective Transform to get the front view of image taken from arbitrary camera angle, and Hough Transform to determine the pointer position. The test results verify the good robustness of the recognition system to the brightness and camera angle.

An Efficient and Effective Automatic Recognition System for Online Recognition of Foreign Fibers in Cotton

Preventing foreign fibers from being mixed with cotton is essential for producing high-quality cotton textile products. It remains a challenging task to accurately distinguish foreign fibers from cotton. This paper proposes an efficient recognition system to accurately recognize foreign fibers mixed in cotton. The core component of the proposed system is an efficient classifier based on the kernel extreme learning machine (KELM). A two-step grid search strategy, which integrates a coarse search with a fine search, is adopted to train an optimal KELM recognition model. The resultant model is compared with the support vector machine and extreme learning machine on a real data set using tenfold cross-validation analysis. The experimental results show that the proposed recognition system can achieve classification accuracy as high as 93.57%, which is superior to the other two state-of-the-art systems. The results clearly confirm the superiority of the developed model in terms of classification accuracy. Promisingly, the proposed system can serve as a new candidate of powerful foreign fiber recognition systems with excellent performance.

Glottal flow 신호에서의 향상된 특징추출 및 다중 특징파라미터 결합을 통한 화자인식 성능 향상

In this paper, we utilize source mel-frequency cepstral coefficients (SMFCCs), skewness, and kurtosis extracted in glottal flow signals to improve speaker recognition performance. Generally, because the high band magnitude response of glottal flow signals is somewhat flat, the SMFCCs are extracted using the response below the predefined cutoff frequency. The extracted SMFCC, skewness, and kurtosis are concatenated with conventional feature parameters. Then, dimensional reduction by the principal component analysis (PCA) and the linear discriminat analysis (LDA) is followed to compare performances with conventional systems under equivalent conditions. The proposed recognition system outperformed the conventional system for large scale speaker recognition experiments. Especially, the performance improvement was more noticeable for small Gaussan mixtures.

Melanoma recognition using extended set of descriptors and classifiers

The paper presents a novel method of melanoma recognition on the basis of dermoscopic images. We use color images of skin lesions, advanced image processing, and different classifiers to distinguish melanoma from the other non-melanoma lesions. Different families of descriptors are used for generation of the image diagnostic features for final pattern recognition. To increase the efficiency of the system, we apply different selection procedures to find the best set of features and different solutions of classifier. The numerical results concerning the accuracy of the proposed recognition system have confirmed good accuracy of the proposed method and high sensitivity in melanoma recognition.

Robust palmprint recognition based on the fast variation Vese–Osher model

Palmprint is usually captured with a touchless device. Due to the changes of angle and position of the palm in a capturing process, as well as the defocus of device, it is inevitable to have some distortions in translation, rotation, and image blurriness, which would degrade the performance of a palmprint recognition system. In order to effectively solve the low recognition problem due to image blurriness, we propose a robust palmprint recognition system in this paper by using the fast Vese–Osher decomposition model to process the blurred palmprint images. First, a Gaussian defocus degradation model (GDDM) is proposed to characterize the image blurriness, and we can observe from this model that there are some stable features in the palmprint images with different scale of blurriness. Second, the structure layer and texture layer of blurred palmprint images are obtained by using the fast Vese–Osher decomposition model, and the structure layer is proved to be more stable and robust than the texture layer for palmprint recognition. Consequently, a novel algorithm based on the weighted histogram of oriented gradient for locally selected pattern (WHOG-LSP) is proposed and it is used to extract some robust features from the structure layer of blurred palmprint images. These extracted features can be used to address low performance issues associated with translation and rotation in palmprint recognition. Finally, the normalized correlation coefficient (NCC) is used to measure the similarity of palmprint features for the proposed recognition system. Extensive experiments on the PolyU palmprint database and the blurred PolyU palmprint database validate the effectiveness and real-time applicability of the proposed recognition system. In addition, an experiment is carried out in the IIT Delhi Touchless Palmprint Database for verifying the robustness of the proposed method.

Optimized projections for nonnegative linear reconstruction classification

Nonnegative linear reconstruction measure can produce the sparse nonnegative representation coefficients which are concentrated on the similar training samples. The samples with the same class-label are generally more similar than those with different class-labels. Based on this, we develop a nonnegative linear reconstruction based classifiers (NLRC). To enhance the classification performance, we design a feature extractor based on the same decisional rule of NLRC, such that the extracted features are most suitable to NLRC. The proposed feature extractor, nonnegative linear reconstruction projection (NLRP), and NLRC share a common decisional rule, and thus the features extracted by NLRP fit NLRC well in theory. The feasibility and effectiveness of the proposed recognition system are verified on the Yale face database, the AR face database the PolyU finger knuckle print database, and the palmprint database with promising results.

Automated inspection of micro-defect recognition system for color filter

This study focused on micro-defect recognition and classification in color filters. First, six types of defects were examined, namely grain, black matrix hole (BMH), indium tin oxide (ITO) defect, missing edge and shape (MES), highlights, and particle. Orthogonal projection was applied to locate each pixel in a test image. Then, an image comparison was performed to mark similar blocks on the test image. The block that best resembled the template was chosen as the new template (or matching adaptive template). Afterwards, image subtraction was applied to subtract the pixels at the same location in each block of the test image from the matching adaptive template. The control limit law employed logic operation to separate the defect from the background region. The complete defect structure was obtained by the morphology method. Next, feature values, including defect gray value, red, green, and blue (RGB) color components, and aspect ratio were obtained as the classifier input. The experimental results showed that defect recognition could be completed as fast as 0.154s using the proposed recognition system and software. In micro-defect classification, back-propagation neural network (BPNN) and minimum distance classifier (MDC) served as the defect classification decision theories for the five acquired feature values. To validate the proposed system, this study used 41 defects as training samples, and treated the feature values of 307 test samples as the BPNN classifier inputs. The total recognition rate was 93.7%. When an MDC was used, the total recognition rate was 96.8%, indicating that the MDC method is feasible in applying automatic optical inspection technology to classify micro-defects of color filters. The proposed system is proven to successfully improve the production yield and lower costs.

瞼の形状近似を利用した操作インタフェースの開発

This paper describes a new method of operating the page turner machine with gazing input. User's face was observed by the USB camera, and it was taken to the control computer to recognize the user's gazing directions. Gazing points with respect to right or left directions could be recognized by the relative distance of an iris and the outer corner of left and right eyes. The states of gazing the camera and reading a book were judged by the curvature of the eyelid shape which was approximated by the Bezier curve. As the gazing point became lower than the height of the camera, the curvature value of the approximation curve was decreased because the eyelid shape roughly became linear. The proposed recognition system was applied to the operations of the page turner machine. In the test trial, it could be confirmed that the developed system could recognize the user's intention, and the subjects were consequently able to operate the page turner machine by only gazing input with non-contact.

Neural Gen Feature Selection for Supervised Learning Classifier

Face recognition has recently received significant attention, especially during the past few years. Many face recognition techniques were developed such as PSO-SVM and LDA-SVM However, inefficient features in the face recognition may lead to inadequate in the recognition results. Hence, a new face recognition system based on Genetic Algorithm and FFBNN technique is proposed. Our proposed face recognition system initially performs the feature extraction and these optimal features are promoted to the recognition process. In the feature extraction, the optimal features are extracted from the face image database by Genetic Algorithm (GA) with FFBNN and the computed optimal features are given to the FFBNN technique to carry out the training and testing process. The optimal features from the feature database are fed to the FFBNN for accomplishing the training process. The well trained FFBNN with the optimal features provide the recognition result. The optimal features in FFBNN by GA efficiently perform the face recognition process. The human face dataset called YALE is utilized to analyze the performance of our proposed GA-FFNN technique and also this GA-FFBNN is compared with standard SVM and PSO-SVM techniques.

A Face Recognition Scheme Based On Principle Component Analysis and Wavelet Decomposition

In this paper, a new face recognition system based on Wavelet transform (HWT) and Principal Component Analysis (PCA) is presented. The image face is preprocessed and detected. The Haar wavelet is used to form the coefficient matrix for the detected face. The image feature vector is obtained by computing PCA for the coefficient matrix of DWT. A comparison between the proposed recognition system using DWT, PCA and Discrete Cosine Transform (DCT) is also made.

An Adaptive Iris Recognition System with Aid of Local Histogram and Optimized FFBNN-AAPSO

Iris recognition is the process of recognizing a person by analyzing the apparent pattern of his or her iris. Many techniques have been developed for iris recognition so far. A method that has been recently developed uses local histogram and image statistics. But it failed to locate the inner and the outer boundaries of the iris in the presence of dense eyelashes which results in poor pupil and iris localization. It also has high computational complexity which reduces the performance of the system. Here we propose a new iris recognition system with the help of local histogram and optimized with FFBNN-AAPSO. In the proposed system, first the input eye images are fetched from the iris database, it is then preprocessed using adaptive median filter to remove noise. Then the features which are extracted from the preprocessed image are given to FFBNN for training the neural network. In order to get accurate results, the FFBNN parameters are optimized using the proposed AAPSO (Adaptive Acceleration Particle Swarm Optimization). In the testing process, the images are preprocessed and subjected to feature extraction process. Then the output obtained from the feature extraction process is given to well trained and optimized FFBNN-AAPSO to check whether the given image is recognized or not. In order to analyze the performance of our proposed recognition system, images from UBRIS iris database are used and the performance of the system is compared with an existing system, PSO-Feed Forward Neural Network and Feed Forward NN.

Walsh, Texture and GIST Descriptors with Bayesian Networks for Recognition of Tifinagh Characters

his paper provides an approach to automatically recognize the Tifinagh characters. The proposed recognition system is based on Texture, Walsh transformation and GIST descriptors as feature extraction methods while the Bayesian Networks are used as a classifier. A comparative study between the Texture descriptor, Walsh transformation and GIST descriptor is given. The experimental results are obtained using a character database of isolated Amazigh characters. A recognition rate of 98.18% is achieved using GIST descriptors.