Enhanced Palmprint Identification Using Score Level Fusion

Collagen is the most prominent protein in the human body, making up 30% of the total protein content. Quantitative studies have shown structural differences between collagen fibers of the normal and diseased tissues, due to the remodeling of the extracellular matrix during the pathological process. The dominant orientation, which is an important characteristic of collagen fibers, has not been taken into consideration for quantitative collagen analysis. Based on the conventional gray level co-occurrence matrix (GLCM) method, the authors proposed the orientation-dependent GLCM (OD-GLCM) method by estimating the dominant orientation of collagen fibers. The authors validated the utility of the OD-GLCM method on second harmonic generation (SHG) microscopic images of tendons from rats with different ages. Compared with conventional GLCM method, the authors' method has not only improved the discrimination between different tissues but also provided additional texture information of the orderliness of collagen fibers and the fiber size. The OD-GLCM method was further applied to the differentiation of the preliminary SHG images of normal and cancerous human pancreatic tissues. The combination of SHG microscopy and the OD-GLCM method might be helpful for the evaluation of diseases marked with abnormal collagen morphology.

Machine learning methods are used to detect elderly accidents while image processing is used to support machine learning performance so that detection performance can be better. This study used two methods used simultaneously, namely GLCM and ANN. The study consisted of preparation of human gesture datasets, preprocessing stage, application of GLCM, analysis of feature extraction results, classification using ANN and analysis of motion class detection results. Overall, the GLCM method with homogeneity parameters and ANN as a classifier obtained an accuracy of 24.32%. The GLCM method with contrast parameters and ANN as a classifier gets an accuracy of 99.84%. The GLCM method with mean parameters and ANN as the classifier gets an accuracy of 99.99%. The GLCM method with dissimilarity parameters and ANN as a classifier to classify hand movement images gets the best accuracy of 100%.

Speaker verification based on the fusion of speech acoustics and inverted articulatory signals

With the increasing demand of highly accurate and robust authentication system, palm print has widely been used in personal identification. Multispectral palm print has been proposed to get more distinguishing feature information and improve resistance to spoof. The vivid texture information of palm print presenting at different resolutions offers abundant prospects in personal recognition. So this paper describes a method to authenticate individuals based on multi-scale and multi-resolution palm print recognition. First, we extract palm print feature based on NSCT (nonsubsampled Contour transform), and store the feature using a hash table. Then the scores generated from each set of palm print image under red, green, blue and NIR spectrum, are combined using score level fusion method. This method use SUM and MAX operators. Comparatively low values of equal error rate and high recognition rate have been obtained for all fusion techniques. Multispectral palm print verification results on the ROI image libraries of PolyU demonstrate the effectiveness and accuracy of the proposed method.

The major hurdle for segmenting lung lobes in computed tomographic (CT) images is to identify fissure regions, which encase lobar fissures. Accurate identification of these regions is difficult due to the variable shape and appearance of the fissures, along with the low contrast and high noise associated with CT images. This paper studies the effectiveness of two texture analysis methods - the gray level co-occurrence matrix (GLCM) and the gray level run length matrix (GLRLM) - in identifying fissure regions from isotropic CT image stacks. To classify GLCM and GLRLM texture features, we applied a feed-forward back-propagation neural network and achieved the best classification accuracy utilizing 16 quantized levels for computing the GLCM and GLRLM texture features and 64 neurons in the input/hidden layers of the neural network. Tested on isotropic CT image stacks of 24 patients with the pathologic lungs, we obtained accuracies of 86% and 87% for identifying fissure regions using the GLCM and GLRLM methods, respectively. These accuracies compare favorably with surgeons/radiologists' accuracy of 80% for identifying fissure regions in clinical settings. This shows promising potential for segmenting lung lobes using the GLCM and GLRLM methods.

The paper presents performance comparison of palm print identification techniques based on fractional coefficients of transformed palm print image using three different transforms like Sine, Walsh and Slant. In transform domain, the energy of image gets accumulated towards high frequency region; this characteristic of image transforms is exploited here to reduce the feature vector size of palm print images by selecting these high frequency coefficients in transformed palm print images. Three image transforms and 12 ways of taking fractional coefficients result into total 36 palm print identification methods. A database of 2350 palm print images is used as a test bed for performance comparison of the proposed palm print identification methods with help of false acceptance rate (FAR) and genuine acceptance rate (GAR). The experimental results in Sine and Walsh transform have shown performance improvement in palm print identification using fractional coefficients of transformed images. In all Sine transform at fractional coefficients of 0.78% gives best performance as indicated by higher GAR value. Thus the task of speeding up palm print identification with better performance is achieved to make it more suitable for real time applications.KeywordsPalm PrintBiometricSine TransformWalsh TransformSlant Transform

Image analysis-based classification of pigmentation patterns in fish: A case study of pseudo-albinism in Senegalese sole

This aim of this study was to assess the discriminatory value of fractal and grey level co-occurrence matrix (GLCM) analysis methods in standard microscopy analysis of two histologically similar brain white mass regions that have different nerve fiber orientation. A total of 160 digital micrographs of thionine-stained rat brain white mass were acquired using a Pro-MicroScan DEM-200 instrument. Eighty micrographs from the anterior corpus callosum and eighty from the anterior cingulum areas of the brain were analyzed. The micrographs were evaluated using the National Institutes of Health ImageJ software and its plugins. For each micrograph, seven parameters were calculated: angular second moment, inverse difference moment, GLCM contrast, GLCM correlation, GLCM variance, fractal dimension, and lacunarity. Using the Receiver operating characteristic analysis, the highest discriminatory value was determined for inverse difference moment (IDM) (area under the receiver operating characteristic (ROC) curve equaled 0.925, and for the criterion IDM≤0.610 the sensitivity and specificity were 82.5 and 87.5%, respectively). Most of the other parameters also showed good sensitivity and specificity. The results indicate that GLCM and fractal analysis methods, when applied together in brain histology analysis, are highly capable of discriminating white mass structures that have different axonal orientation.

Multimodal biometrics-based systems aim to improve the recognition accuracy of human beings using more than one physical and/or behavioral characteristics of a person. In this paper, different fusion schemes at matching score level and feature level are employed to obtain a robust recognition system using several standard feature extractors. The proposed method involves the consideration of a face–iris multimodal biometric system using score level and feature level fusion. Principal Component Analysis (PCA), subspace Linear Discriminant Analysis (LDA), subpattern-based PCA, modular PCA and Local Binary Patterns (LBP) are global and local feature extraction methods applied on face and iris images. In fact, different feature sets obtained from five local and global feature extraction methods for unimodal iris biometric system are concatenated at feature level fusion called iris feature vector fusion (iris-FVF), while for unimodal face biometric system, LBP is used to achieve efficient texture descriptors. Feature selection is performed using Particle Swarm Optimization (PSO) at feature level fusion step to reduce the dimension of feature vectors for improving the recognition performance. Our proposed method is validated by forming three datasets using ORL, BANCA, FERET face databases and CASIA, UBIRIS iris databases. The results based on recognition performance and ROC analysis demonstrate that the proposed matching score level fusion scheme using Weighted Sum rule, tanh normalization, iris-FVF and facial features extracted by LBP achieves a significant improvement over unimodal and multimodal methods. Support Vector Machine (SVM) and t-norm normalization are also used to improve the recognition performance of the proposed method.

One of the most important problems in pattern recognition is texture-based image classification. In this paper, the combination of Discrete Cosine Transform (DCT) and Gray Level Co-Occurrence Matrix (GLCM) methods for feature extraction was proposed. The attributes extracted from DCT method were mean and variance, while the attributes extracted from GLCM method were energy and entropy. Adaptive Neuro-Fuzzy Inference System (ANFIS) was used as a classifier. The classifier model was trained using 50% of the texture images and remaining images were used for testing. Four classes of texture images were downloaded from KTH-TIPS (Textures under varying Illumination, Pose and Scale) image database, three of which were used in each experiments thus there were four data combination. The best data testing accuracy result towards textures of crumpled aluminium foil, corduroy, and orange peel is 98.3%, which is 1.6% better than one hidden-layer feed forward neural network classifier. In average, testing accuracy result of ANFIS excelled one hidden-layer feed forward neural network with 93.7% over 90.4%.

This paper presents a new personal authentication system that simultaneously exploits 2D and 3D palm print and geometrical features of the hand. The geometrical features are obtained from the binaries images by using Otsu’s threshold technique. Multimodal (2-D and 3D) palm print and hand geometry features, which are simultaneously extracted from the user’s textured 2-D and 3-D hand, are used for matching. Individual matching scores are then combined using a new Hybrid level fusion strategy. The objective of this work is to improve accuracy and robustness of existing palm print authentication systems using combination of the 2D and 3D palm print features. A peg-free system is composed by a pc and camera. The users put the hand in free space in front of the camera. The hand is illuminated by an infra- red light to solve segmentation problems in a real environment. The surface curvature feature based method is investigated for Gabor feature based competitive coding scheme is used for 2D representation. The database of 120 subjects achieved significant improvement in performance with the integration of 2D and 3D palm print and hand geometry features. Such as those mounted on a laptop, mobile device, and web camera or those for surveillance, can dramatically increase the applicability of such a system. However, the performance of existing techniques for palm print authentication falls considerably, when the camera is not aligned with the surface of the palm. The experimental results also suggest that the Hybrid level fusion approach employed in this work helps to achieve the performance improvement of 70% (in terms of EER) over the case when matching scores are combined using the Dynamic fusion approach.

Face recognition system contains lots of challenges due to various environmental factors, background variations, poor quality of camera, different illumination and others. Since twins are involved with criminal activities, twin identification becomes an essential task. The proposed system is focused on identifying the identical twins for the still images. The fusion based approach has been implemented in the proposed system. It combines the features extracted by using Principal Component Analysis (PCA), Histogram Oriented of Gradients (HOG), Local Binary Pattern (LBP), Gabor and distance between the facial components. Three types of fusion such as Decision Level Fusion, Feature Level Fusion and Score Level Fusion are used in the proposed approach. Based on these fusion generated scores, the twin has been identified. In the proposed system, Particle Swarm Optimization is used for the best feature selection and SVM classifier is used for training and testing the image. The proposed system provides better results when compared with the other twin detection techniques.

Breast cancer is one of the main causes of death in women and ranks first in cancer cases in Indonesia. Therefore, an early detection and prevention of breast cancer is necessary, one of which is through mammography procedures. A machine learning classifier such as Support Vector Machines (SVM) could be used as an aid to the doctors and radiologist in diagnosing breast cancer from the mammogram images. The aim of this paper is to compare two feature extraction methods used in SVM, namely the Gray Level Co-Occurrence Matrix (GLCM) and first order with two kernels for each method, namely Gaussian and Polynomial. Classification using SVM method is carried out by testing several parameters such as the value of C, gamma, degree and varying the pixel spacing values ​​in GLCM, which usually in previous studies only used the default pixel spacing. The dataset consists of 500 mammogram images containing 250 benign and malignant images, respectively. This study is expected to find out the best method with the highest accuracy between these two texture feature extractions and and able to distinguish between benign and malignant classes correctly. The result achieved that Gray Level Co-Occurrence Matrix (GLCM) feature extraction method with both Gaussian and Polynomial kernel yields the best performance with an accuracy of 89%.

본 논문에서는 비균일 양자화에 기반을 둔 영상의 질감분석에 널리 이용되고 있는 gray level co-occurrence matrix(GLCM)의 성능개선을 제안하였다. 여기서 비균일 양자화는 평균자승오차의 최소화를 위한 반복계산 기법인 Lloyd 알고리즘을 이용하였다. 이는 영상에서의 비균일 양자화 과정으로 얻어지는 비선형의 명암레벨을 GLCM의 생성에 이용함으로써 행렬의 차원을 감소시켜, GLCM의 생성과 질감특성 파라미터들의 계산에 따른 부하를 줄이기 위함이다. 제안된 기법을 30개의 <TEX>$120{\times}120$</TEX> 픽셀의 256 그레이 레벨을 가진 영상들을 대상으로 적용하여 angular second moment, contrast, variance, entropy, correlation, inverse difference moment 6개의 질감특성 파라미터들을 각각 계산한 실험결과, 양자화를 수행하지 않은 256 레벨 GLCM에 비해 계산시간과 저장 공간에서 개선된 성능이 있음을 확인하였다. 특히 48, 32, 16, 12, 8의 비균일 양자화 레벨 중에서 16일 때 가장 우수한 질감특성분석 성능이 있음을 알 수 있었다. This paper presents a performance improvement of gray level co-occurrence matrix(GLCM) based on the nonuniform quantization, which is generally used to analyze the texture of images. The nonuniform quantization is given by Lloyd algorithm of recursive technique by minimizing the mean square error. The nonlinear intensity levels by performing nonuniformly the quantization of image have been used to decrease the dimension of GLCM, that is applied to reduce the computation loads as a results of generating the GLCM and calculating the texture parameters by using GLCM. The proposed method has been applied to 30 images of <TEX>$120{\times}120$</TEX> pixels with 256-gray level for analyzing the texture by calculating the 6 parameters, such as angular second moment, contrast, variance, entropy, correlation, inverse difference moment. The experimental results show that the proposed method has a superior computation time and memory to the conventional 256-level GLCM method without performing the quantization. Especially, 16-gray level by using the nonuniform quantization has the superior performance for analyzing textures to another levels of 48, 32, 12, and 8 levels.

In modern medicine, image recognition via segmentation of anatomical regions and automatic classification of diseases using medical images has a growing potential role in diagnosis of various diseases. Scintigraphy of thyroid is one of the established imaging modalities for diagnosis of thyroid gland disorders. In our study, the speckle noise was reduced in the scintigraphy images with the optimized Bayesian nonlocal mean filter. The thyroid gland was automatically segmented by local based active contour method and the thyroid gland pathologies were classified with convolutional neural networks (CNN). The proposed computer aided diagnosis (CAD) system was compared with Pyramid of Histograms of Orientation Gradients (PHOG), Gray Level Co occurrence Matrix (GLCM), Local Configuration Pattern (LCP) and Bag of Feature (BoF) methods. The common pathological patterns of scintigraphic images of the thyroid gland were successfully classified by CNN with an overall success rate of 91.19%. The comparative methods were PHOG, GLCM, LCP and BoF methods which provided overall success rates of 7.61%, 86.04%, 88.91% and 85.72% respectively. The proposed CNN based automatic diagnosis system provided promising results compared to handcrafted methods.