Tarek Gaber

Biometric cattle identification approach based on Weber's Local Descriptor and AdaBoost classifier

Limitations and weakness of traditional cattle identification were highlighted.Cattle identification approach based on biometric features was proposed.Webers Local Descriptor (WLD) along with many classifiers were examined.Dataset of muzzle print images (images from 31 cattle animals) were used.The results were validated against False Accept and Reject rate, Sensitivity and Specificity, and accuracy. In this paper, we proposed a new and robust biometric-based approach to identify head of cattle. This approach used the Weber Local Descriptor (WLD) to extract robust features from cattle muzzle print images (images from 31 head of cattle were used). It also employed the AdaBoost classifier to identify head of cattle from their WLD features. To validate the results obtained by this classifier, other two classifiers (k-Nearest Neighbor (k-NN) and Fuzzy-k-Nearest Neighbor (Fk-NN)) were used. The experimental results showed that the proposed approach achieved a promising accuracy result (approximately 99.5%) which is better than existed proposed solutions. Moreover, to evaluate the results of the proposed approach, four different assessment methods (Area Under Curve (AUC), Sensitivity and Specificity, accuracy rate, and Equal Error Rate (EER)) were used. The results of all these methods showed that the WLD along with AdaBoost algorithm gave very promising results compared to both of the k-NN and Fk-NN algorithms.

Cattle Identification Based on Muzzle Images Using Gabor Features and SVM Classifier

The accuracy of animal identification plays an important role for producers to make management decisions about their individual animal or about their complete herd. The animal identification is also important to animal traceability systems as ensure the integrity of the food chain. Usually, recording and reading of tags-based systems are used to identify animal, but only effective in eradication programs of national disease. Recently, animal biometric-based solutions, e.g. muzzle imaging system, offer an effective and secure, and rapid method of addressing the requirements of animal identification and traceability systems. In this paper, we propose a robust and fast cattle identification through using Gabor filter-based feature extraction method. We extract Gabor features from three different scales of muzzle print images. SVM classifier with its different kernels (Gaussian, Polynomial, Linear and Sigmoid) has been applied to Gabor features. Also, two different levels of fusion are used namely feature fusion and classifier fusion. The experimental results showed that Gaussian-based SVM classifier has achieved the best accuracy among all other kernels and generally our approach is superior than existed works as ours achieves 99.5% identification accuracy. In addition, the identification rate when the fusion is done at the feature level is better than that is done at classification level.

SIFT-Based Arabic Sign Language Recognition System

The literature contains many proposed solutions for automatic sign language recognition. However, the ArSL (Arabic Sign Language), unlike ASL (American Sign Language), did not take much attention from the research community. In this paper, we propose a new system which does not require a deaf wear inconvenient devices like gloves to simplify the process of hand recognition. The system is based on gesture extracted from 2D images. The Scale Invariant Features Transform (SIFT) technique is used to achieve this task as it extracts invariant features which are robust to rotation and occlusion. Also, the Linear Discriminant Analysis (LDA) technique is used to solve dimensionality problem of the extracted feature vectors and to increase the separability between classes, thus increasing the accuracy of the introduced system. The classifiers, Support Vector Machine (SVM), k-Nearest Neighbor (k-NN), and minimum distance will be used to identify the Arabic sign characters. Experiments are conducted to check the performance of the proposed system and it showed that the accuracy of the obtained results is around 99%. Also, the experiments proved that the proposed system is robust against any rotation and they achieved an identification rate near to 99%. Moreover, the evaluation shown that the system is comparable to the related work.

Cattle Identification Using Muzzle Print Images Based on Texture Features Approach

The increasing growth of the world trade and growing concerns of food safety by consumers need a cutting-edge animal identification and traceability systems as the simple recording and reading of tags-based systems are only effective in eradication programs of national disease. Animal biometric-based solutions, e.g. muzzle imaging system, offer an effective and secure, and rapid method of addressing the requirements of animal identification and traceability systems. In this paper, we propose a robust and fast cattle identification approach. This approach makes use of Local Binary Pattern (LBP) to extract local invariant features from muzzle print images. We also applied different classifiers including Nearest Neighbor, Naive Bayes, SVM and KNN for cattle identification. The experimental results showed that our approach is superior than existed works as ours achieves 99,5% identification accuracy. In addition, the results proved that our proposed method achieved this high accuracy even if the testing images are rotated in various angels or occluded with different parts of their sizes.

Linear discriminant analysis: A detailed tutorial

Linear Discriminant Analysis (LDA) is a very common technique for dimensionality reduction problems as a preprocessing step for machine learning and pattern classification applications. At the same time, it is usually used as a black box, but (sometimes) not well understood. The aim of this paper is to build a solid intuition for what is LDA, and how LDA works, thus enabling readers of all levels be able to get a better understanding of the LDA and to know how to apply this technique in different applications. The paper first gave the basic definitions and steps of how LDA technique works supported with visual explanations of these steps. Moreover, the two methods of computing the LDA space, i.e. class-dependent and class-independent methods, were explained in details. Then, in a step-by-step approach, two numerical examples are demonstrated to show how the LDA space can be calculated in case of the class-dependent and class-independent methods. Furthermore, two of the most common LDA problems (i.e. Small Sample Size (SSS) and non-linearity problems) were highlighted and illustrated, and state-of-the-art solutions to these problems were investigated and explained. Finally, a number of experiments was conducted with different datasets to (1) investigate the effect of the eigenvectors that used in the LDA space on the robustness of the extracted feature for the classification accuracy, and (2) to show when the SSS problem occurs and how it can be addressed.

Plant Identification: Two Dimensional-Based Vs. One Dimensional-Based Feature Extraction Methods

In this paper, a plant identification approach using 2D digital leaves images is proposed. The approach made use of two methods of features extraction (one-dimensional (1D) and two-dimensional (2D) techniques) and the Bagging classifier. For the 1D-based method, PCA and LDA techniques were applied, while 2D-PCA and 2D-LDA algorithms were used for the 2D-based method. To classify the extracted features in both methods, the Bagging classifier, with the decision tree as a weak learner, was used. The proposed approach, with its four feature extraction techniques, was tested using Flavia dataset which consists of 1907 colored leaves images. The experimental results showed that the accuracy and the performance of our approach, with the 2D-PCA and 2D-LDA, was much better than using the PCA and LDA. Furthermore, it was proven that the 2D-LDA-based method gave the best plant identification accuracy and increasing the weak learners of the Bagging classifier leaded to a better accuracy. Also, a comparison with the most related work showed that our approach achieved better accuracy under the same dataset and same experimental setup.

A New Multi-layer Perceptrons Trainer Based on Ant Lion Optimization Algorithm

In this paper, Ant Lion Optimizer (ALO) was presented to train Multi-Layer Perceptron (MLP). ALO was used to find the weights and biases of the MLP to achieve a minimum error and a high classification rate. Four standard classification datasets were used to benchmark the performance of the proposed method. In addition, the performance of the proposed method were compared with three well-known optimization algorithms, namely, Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Ant Colony Optimization (ACO). The experimental results showed that the ALO algorithm with the MLP was very competitive as it solved the local optima problem and achieved a high accuracy rate.

Plants Identification Using Feature Fusion Technique and Bagging Classifier

In this paper, a plant identification approach using 2D digital images of leaves is proposed. This approach will be used to develop an expert system to identify plant species by processing colored images of its leaf. The approach made use of feature fusion technique and the Bagging classifier. Feature fusion technique is used to combine color, shape, and texture features. Color moments, invariant moments, and Scale Invariant Feature Transform (SIFT) are used to extract the color, shape, and texture features, respectively. Linear Discriminant Analysis (LDA) is used to reduce the number of features and Bagging ensemble is used to match the unknown image and the training or labeled images. The proposed approach was tested using Flavia dataset which consists of 1907 colored images of leaves. The experimental results showed that the accuracy of feature fusion approach was much better than all other single features. Moreover, a comparison with the most related work showed that our approach achieved better accuracy under the same dataset and same experimental setup.

Detection of breast abnormalities of thermograms based on a new segmentation method

Breast cancer is one from various diseases that has got great attention in the last decades. This due to the number of women who died because of this disease. Segmentation is always an important step in developing a CAD system. This paper proposed an automatic segmentation method for the Region of Interest (ROI) from breast thermograms. This method is based on the data acquisition protocol parameter (the distance from the patient to the camera) and the image statistics of DMR-IR database. To evaluated the results of this method, an approach for the detection of breast abnormalities of thermograms was also proposed. Statistical and texture features from the segmented ROI were extracted and the SVM with its kernel function was used to detect the normal and abnormal breasts based on these features. The experimental results, using the benchmark database, DMR-IR, shown that the classification accuracy reached (100%). Also, using the measurements of the recall and the precision, the classification results reached 100%. This means that the proposed segmentation method is a promising technique for extracting the ROI of breast thermograms.

Thermogram breast cancer prediction approach based on Neutrosophic sets and fuzzy c-means algorithm.

The early detection of breast cancer makes many women survive. In this paper, a CAD system classifying breast cancer thermograms to normal and abnormal is proposed. This approach consists of two main phases: automatic segmentation and classification. For the former phase, an improved segmentation approach based on both Neutrosophic sets (NS) and optimized Fast Fuzzy c-mean (F-FCM) algorithm was proposed. Also, post-segmentation process was suggested to segment breast parenchyma (i.e. ROI) from thermogram images. For the classification, different kernel functions of the Support Vector Machine (SVM) were used to classify breast parenchyma into normal or abnormal cases. Using benchmark database, the proposed CAD system was evaluated based on precision, recall, and accuracy as well as a comparison with related work. The experimental results showed that our system would be a very promising step toward automatic diagnosis of breast cancer using thermograms as the accuracy reached 100%.