Pyramid Histogram Of Oriented Gradients Research Articles

Flamingo Search Sailfish Optimizer Based SqueezeNet for Detection of Breast Cancer Using MRI Images

Breast cancer with increased risk in women is identified with Breast Magnetic Resonance Imaging (Breast MRI) and this helps in evaluating treatment therapies. Breast MRI is time time-consuming process that involves the assessment of current imaging. This research work depends on the detection of breast cancer at the earlier stages. Among various cancers, breast cancer in women occurs in larger accounts for almost 30% of estimated cancer cases. In this research, many steps are followed for breast cancer detection like pre-processing, segmentation, augmentation, extraction of features, and cancer detection. Here, the median filter is utilized for pre-processing, as well as segmentation is followed after pre-processing, which is done by Psi-Net. Moreover, the process of augmentation like shearing, translation, and cropping are followed after segmentation. Also, the segmented image tends to process feature extraction, where features like shape features, Completed Local Binary Pattern (CLBP), Pyramid Histogram of Oriented Gradients (PHOG), and statistical features are extracted. Finally, breast cancer is detected using the DL model, SqueezeNet. Here, the newly devised Flamingo Search SailFish Optimizer (FSSFO) is used in training Psi-Net as well as SqueezeNet. Furthermore, FSSFO is the combination of both the Flamingo Search Algorithm (FSA) and SailFish Optimizer (SFO).

Adam golden search optimization enabled DCNN for classification of breast cancer using histopathological image

Breast Cancer (BC) is a killing disorder, every year it kills millions of human beings. Early diagnosis is the only way to mitigate the mortality rate. Among all kinds of screening methods, medical imaging is an essential method for screening BC. Existing medical imaging alters the tissue structure and cell morphology. To overcome these limitations, histopathology image is used because it can support the decision of pathologists about the closeness or the non-appearance of a disease, as well as it can help in infection development estimation. Hence, this research develops an efficient method for BC classification using the proposed Adam Golden Search Optimization-based Deep Convolutional Neural Network (AGSO-DCNN). Initially, Gaussian filter-enabled pre-processing is utilized for mitigating the noises composed in the input images. Afterwards, k-means clustering is used to feed the input images into the segmentation phase to reduce the complexity of the image. Then, to extract features like shape features, statistical features, Local Vector Patterns (LVP), and Pyramid Histogram of Oriented Gradients (PHOG) feature extraction is performed. Thereafter, the obtained features are forwarded to the multi-grade BC classification stage, where DCNN is employed for classifying the image into six categories, such as apoptosis, tubule, mitosis, non-tubule, tumour nuclei, and non-tumor nuclei. DCNN is trained by the formulated AGSO mechanism, which is obtained by incorporating the Adam Optimizer and Golden Search Optimization (GSO) algorithm. Moreover, the AGSO-based DCNN technique achieved better accuracy, TPR and TNR with the values of 97.90%, 98.00%, and 98.30%, respectively.

Microscopic parasite malaria classification using best feature selection based on generalized normal distribution optimization.

Malaria disease can indeed be fatal if not identified and treated promptly. Due to advancements in the malaria diagnostic process, microscopy techniques are employed for blood cell analysis. Unfortunately, the diagnostic process of malaria via microscopy depends on microscopic skills. To overcome such issues, machine/deep learning algorithms can be proposed for more accurate and efficient detection of malaria. Therefore, a method is proposed for classifying malaria parasites that consist of three phases. The bilateral filter is applied to enhance image quality. After that shape-based and deep features are extracted. In shape-based pyramid histograms of oriented gradients (PHOG) features are derived with the dimension of N × 300. Deep features are derived from the residual network (ResNet)-50, and ResNet-18 at fully connected layers having the dimension of N × 1,000 respectively. The features obtained are fused serially, resulting in a dimensionality of N × 2,300. From this set, N × 498 features are chosen using the generalized normal distribution optimization (GNDO) method. The proposed method is accessed on a microscopic malarial parasite imaging dataset providing 99% classification accuracy which is better than as compared to recently published work.

Circle-Inspired Sine Cosine Optimization-Enabled CRF-RNN and ZFNet for Brain Tumor Segmentation and Classification Using MRI Images

The segmentation and classification of brain tumor are a attractive regions, which distinguish tumor as well as nontumor cells for identifying a level of tumor. Segmentation and classification from MRI images is a great challenge due to their altering image sizes and vast databases. Various schemes are designed for the segmentation and classification of the brain tumor, but these methods failed to offer better classification accuracy and decision-making. Here, Circle-inspired sine cosine Optimization_conditional random field-recurrent neural network (CISCO-CRF-RNN) and CISCO-Zeiler and Fergus network (CISCO-ZFNet) are introduced for brain tumor segmentation and classification. A pre-processing phase in this research is executed by the median filter to eliminate noises from an image. The segmentation is done by CRF-RNN, which is trained by CISCO. Furthermore, CISCO is newly introduced by incorporation of Circle-Inspired Optimization Algorithm (CIOA) and Sine Cosine Algorithm (SCA). Thereafter, image augmentation is performed utilizing some image augmentation techniques, and thereafter, features namely statistical features, Convolutional Neural Network (CNN) features, haralick features, pyramid histogram of orientation gradients (PHoG), and Local Vector Pattern (LVP) are extracted. Finally, the classification of brain tumor is accomplished utilizing ZFNet, which is tuned using CISCO. In addition, CISCO-CRF-RNN obtained maximal segmentation accuracy of 90.6% whereas CISCO-ZFNet achieved maximum pixel accuracy, negative predictive value (NPV), positive predictive value (PPV), and True positive rate TPR of 92.5%, 89.2%, 90.1% and 93% as well as minimum FNR of 15%.

Analysis of psychological biometric data for gender identification using fusion features and deep learning

A person's gender must be correctly identified for many applications, such as tailored services, social sciences, and human-computer interaction, to function properly. Traditional gender identification techniques mostly depend on physical traits, but new developments in psychological biometrics provide interesting alternatives. Over the last several years, several access control systems have included biometric security technologies to increase security. The handwritten signature is the psychological biometric characteristic that is most often used to validate daily documents like letters, contracts, wills, MOUs, etc. This study proposes an innovative Deep Learning (DL) approach to identify a person's gender from an image of their handwritten signature. The fusion of statistical and textural information taken from the trademark photos serves as the foundation for the proposed work. The texture is represented by the Pyramid Histogram of Oriented Gradients (PHOG) features. A novel sequence labelling multidimensional recurrent neural network (SLMRNN) is employed to classify the writer's gender. Extensive experiments are carried out on the gathered dataset to assess the performance of the suggested technique. The efficacy of the fusion features and DL models for gender recognition is evaluated using a variety of measures, including accuracy, precision, recall, and F1 score. To evaluate the suggested approach against existing procedures and emphasize its advantages, if any, comparative assessments are also carried out. The findings show that, in comparison to conventional approaches, the combination of behavioral biometric variables with cutting-edge DL algorithms greatly enhances gender identification accuracy. The suggested approach is anticipated to be beneficial in the development of effective computer vision tools for forensic analysis and authentication of papers with handwritten signatures.

Application of novel DIRF feature selection algorithm for automated brain disease detection

The brain is a complex organ and Magnetic Resonance Imaging (MRI) is the most widely used imaging modality for diagnosing brain diseases due to its superior soft tissue contrast. In clinical practice, interpreting MRI scans is laborious, time-consuming, and error-prone. While most studies focus on detecting brain tumors and multiple sclerosis, there are limited studies on atrophy, ischemia, and white matter intensity (WMI) diseases. Moreover, most studies are complex, time-consuming, computationally intensive, and specific to the detection of a particular brain disease. Our study presents a new deep learning-based ensemble model and also introduces the novel double iterative ReliefF (DIRF) feature selection algorithm to enhance the performance of automatic detection of atrophy, ischemia, and WMI using MRI images. Two pre-trained deep models, namely AlexNet and GoogleNet, and three well-known local texture descriptors, namely pyramid histogram of oriented gradients (PHOG), binary Gabor pattern (BGP), and binarized statistical image features (BSIF), were employed as feature extractors, and the extracted features were merged. The merged features were fed into the proposed double iterative ReliefF (DIRF) feature selection algorithm, and the most important 688 features were selected. Finally, the selected features were classified by support vector machine (SVM). Our proposed DIRF-based ensemble model achieved 98.87 % accuracy using a 10-fold cross-validation strategy. Our proposed model developed using a public brain diseases MRI dataset achieved an accuracy of 95.96 %. This study demonstrates that our proposed model, with DIRF feature selection algorithm can significantly improve the diagnosis of various brain diseases, enhancing the patient care and outcomes.

Ensemble of Local Texture Descriptor for Accurate Breast Cancer Detection from Histopathologic Images

Histopathological analysis is important for detection of the breast cancer (BC). Computer-aided diagnosis and detection systems are developed to assist the radiologist in the diagnosis process and to relieve the patient from unnecessary pain. In this study, a computer-aided diagnosis system for the early detection of benign and malignant breast cancer is proposed. The proposed system consists of feature extraction, feature ensemble, and classification stages. Various preprocessing steps such as grayscale conversion, noise filtering, and image resizing are employed on the input histopathological images. The local texture descriptors namely Local Binary Pattern (LBP), Frequency Decoded LBP (FDLBP), Binary Gabor Pattern (BGP), Local Phase Quantization (LPQ), Binarized Statistical Image Features (BSIF), CENsus TRansform hISTogram (CENTRIST), and Pyramid Histogram of Oriented Gradients (PHOG) are employed for feature extraction from the histopathologic images. The obtained features are then concatenated for the construction of the ensemble of the features. Three classifiers namely Support Vector Machines (SVM), K-nearest neighbor (KNN), and Neural Networks (NN) are used in the detection of the BC and the classification accuracy score is used for performance evaluation. A dataset called BreaKHis is used in the studies. There are 9109 microscopic pictures in BreaKHis, with 2480 benign samples and 5429 malignant samples. During the collection of the data, 82 patients' breast tumor tissues were envisioned using various magnification factors such as 40X, 100X, 200X, and 400X. The accuracy score is used to assess the acquired findings. The results show that the proposed method has the potential to use accurate BC detection.

Contemporary Advertising Text Art Design and Effect Evaluation by IoT Deep Learning under the Smart City

This work intends to solve the problem that the current artistic typeface generation methods rely too much on manual intervention, lack novelty, and the single font local feature and the global feature extraction method cannot fully describe the font features. Firstly, it proposes a handwritten word recognition model based on generalized search trees (GIST) and the pyramid histogram of oriented gradient (PHOG). The local features and global features of the font are fused. Secondly, a model of automatic artistic typeface generation based on generative adversarial networks (GAN) is constructed, which can use hand-drawn fonts to automatically generate artistic typefaces in the desired style through training as needed. Finally, the generation of the huaniao typeface is used as an example. By constructing the dataset, the effectiveness of the two models is verified. The experimental results show the following: (1) The proposed handwritten character recognition model based on GIST and PHOG has a higher recognition rate of different fonts than the single GIST and PHOG features by more than 5.8%. The total recognition time is reduced by more than 49.4%, and the performance is improved significantly. (2) Compared with other popular algorithms, the constructed GAN-based automatic artistic typeface generation model has the best quality of the generation of huaniao on both the pencil sketch and the calligraphy character image dataset. Models have broad application prospects in contemporary advertising text art design. This study aims to provide important technical support for the automation of contemporary advertising text art design and the improvement of overall efficiency.

CLASSIFICATION OF INFORMATIVE FRAMES IN COLONOSCOPY VIDEO BASED ON IMAGE ENHANCEMENT AND PHOG FEATURE EXTRACTION

Colonoscopy allows doctors to check the abnormalities in the intestinal tract without any surgical operations. The major problem in the Computer-Aided Diagnosis (CAD) of colonoscopy images is the low illumination condition of the images. This study aims to provide an image enhancement method and feature extraction and classification techniques for detecting polyps in colonoscopy images. We propose a novel image enhancement method with a Pyramid Histogram of Oriented Gradients (PHOG) feature extractor to detect polyps in the colonoscopy images. The approach is evaluated across different classifiers, such as Multi-Layer Perceptron (MLP), Adaboost, Support Vector Machine (SVM), and Random Forest. The proposed method has been trained using the publicly available databases CVC ClinicDB and tested in ETIS Larib and CVC ColonDB. The proposed approach outperformed the existing state-of-the-art methods on both databases. The reliability of the classifiers’ performance was examined by comparing their F1 score, precision, F2 score, recall, and accuracy. PHOG with Random Forest classifier outperformed the existing methods in terms of recall of 97.95%, precision 98.46%, F1 score 98.20%, F2 score of 98.00%, and accuracy of 98.21% in the CVC-ColonDB. In the ETIS-LARIB dataset it attained a recall value of 96.83%, precision 98.65%, F1 score 97.73%, F2 score 98.59%, and accuracy of 97.75%. We observed that the proposed image enhancement method with PHOG feature extraction and the Random Forest classifier will help doctors to evaluate and analyze anomalies from colonoscopy data and make decisions quickly.

VGG19 Network Assisted Joint Segmentation and Classification of Lung Nodules in CT Images.

Pulmonary nodule is one of the lung diseases and its early diagnosis and treatment are essential to cure the patient. This paper introduces a deep learning framework to support the automated detection of lung nodules in computed tomography (CT) images. The proposed framework employs VGG-SegNet supported nodule mining and pre-trained DL-based classification to support automated lung nodule detection. The classification of lung CT images is implemented using the attained deep features, and then these features are serially concatenated with the handcrafted features, such as the Grey Level Co-Occurrence Matrix (GLCM), Local-Binary-Pattern (LBP) and Pyramid Histogram of Oriented Gradients (PHOG) to enhance the disease detection accuracy. The images used for experiments are collected from the LIDC-IDRI and Lung-PET-CT-Dx datasets. The experimental results attained show that the VGG19 architecture with concatenated deep and handcrafted features can achieve an accuracy of 97.83% with the SVM-RBF classifier.

Big Cats Classification Based on Body Covering

The reduced habitat owned by an animal has a very bad impact on the survival of the animal, resulting in a continuous decrease in the number of animal populations especially in animals belonging to the big cat family such as tigers, cheetahs, jaguars, and others. To overcome the decline in the animal population, a classification model was built to classify images that focuses on the pattern of body covering possessed by animals. However, in designing an accurate classification model with an optimal level of accuracy, it is necessary to consider many aspects such as the dataset used, the number of parameters, and computation time. In this study, we propose an animal image classification model that focuses on animal body covering by combining the Pyramid Histogram of Oriented Gradient (PHOG) as the feature extraction method and the Support Vector Machine (SVM) as the classifier. Initially, the input image is processed to take the body covering pattern of the animal and converted it into a grayscale image. Then, the image is segmented by employing the median filter and the Otsu method. Therefore, the noise contained in the image can be removed and the image can be segmented. The results of the segmentation image are then extracted by using the PHOG and then proceed with the classification process by implementing the SVM. The experimental results showed that the classification model has an accuracy of 91.07%.

Video summary generation by visual shielding compressed sensing coding and double-layer affinity propagation

Video summary technology based on keyframe extraction is an effective means to rapidly access video content. Traditional video summary generation technology requires high video resolution, which poses a problem as most existing studies have no targeted solutions for videos that are subject to privacy protection. We propose a novel keyframe extraction algorithm for video data in the visual shielding domain, named visual shielding compressed sensing coding and double-layer affinity propagation (VSCS-DAP). VSCS-DAP involves three main steps. First, the video is compressed by compressed sensing technology to provide a visual shielding effect (protecting the privacy of monitored figures), while the data volume is significantly reduced. Then, pyramid histogram of oriented gradients (PHOG) features are extracted from the compressed video to be clustered by the first step affinity propagation (AP) to gain the summaries of the first stage. Finally, the PHOG and Hist fusion features are extracted from the keyframes of the first stage, and they cluster the fused PHOG-Hist features by the second step AP algorithm to obtain the final output summaries. Experimental results obtained on two common video datasets show that our method exhibits advantages including low redundancy and few missing frames, low computational complexity, strong real-time performance, and robustness to vision-shielded video.

An Ensemble Learning-Based Method for Inferring Drug-Target Interactions Combining Protein Sequences and Drug Fingerprints.

Identifying the interactions of the drug-target is central to the cognate areas including drug discovery and drug reposition. Although the high-throughput biotechnologies have made tremendous progress, the indispensable clinical trials remain to be expensive, laborious, and intricate. Therefore, a convenient and reliable computer-aided method has become the focus on inferring drug-target interactions (DTIs). In this research, we propose a novel computational model integrating a pyramid histogram of oriented gradients (PHOG), Position-Specific Scoring Matrix (PSSM), and rotation forest (RF) classifier for identifying DTIs. Specifically, protein primary sequences are first converted into PSSMs to describe the potential biological evolution information. After that, PHOG is employed to mine the highly representative features of PSSM from multiple pyramid levels, and the complete describers of drug-target pairs are generated by combining the molecular substructure fingerprints and PHOG features. Finally, we feed the complete describers into the RF classifier for effective prediction. The experiments of 5-fold Cross-Validations (CV) yield mean accuracies of 88.96%, 86.37%, 82.88%, and 76.92% on four golden standard data sets (enzyme, ion channel, G protein-coupled receptors (GPCRs), and nuclear receptor, respectively). Moreover, the paper also conducts the state-of-art light gradient boosting machine (LGBM) and support vector machine (SVM) to further verify the performance of the proposed model. The experimental outcomes substantiate that the established model is feasible and reliable to predict DTIs. There is an excellent prospect that our model is capable of predicting DTIs as an efficient tool on a large scale.

An efficient facial emotion recognition system using novel deep learning neural network-regression activation classifier

In the computer vision field, FER encompasses a significant place. It is being studied for a long period, and in recent decades, it has attained progress, but all is in vain, since, recognizing facial expression with high accuracy is still hard due to disparate facial expressions. To beat such difficulties, an efficient Facial Emotion Recognitions (FER) is proposed by utilizing a novel Deep Learning Neural Network-regression activation (DR) classifier. The proposed method has six phases, namely, pre-processing, facial point extraction, segmentation, feature extraction, feature selection, and classification. Initially the input image has been pre-processed using Gamma-HE technique and then facial points are extracted using Pyramid Histogram of Oriented Gradients (PHOG) based Supervised Descent (SMD) Method. The facial parts are segmented using Viola-Jones Algorithm (VJA) and then Local Tetra Pattern (LTrP), cluster shade, Inverse Divergent Moment (IDM), Local homogeneity, optimum probability, cluster prominence, dissimilarity, autocorrelation, and contrast features have been extracted. Modified Monarch Butterfly Optimization (MMBO) algorithm has been used to select necessary features from the extracted features. From the extracted facial points, the DR classifier classifies the emotions of the particular input image. A ‘2’ datasets were taken for analyzing the proposed system’s performance. Centred on the CK+ database, the proposed work attains 0.9885-accuracy, and centred on the JAFFE database, it has 0.9727-accuracy. Also, the investigational results proved that the proposed work trounces the existing systems centred on statistical metrics.

A novel method for vehicle headlights detection using salient region segmentation and PHOG feature

In this paper, we explore an issue that is to detect vehicle headlights from the nighttime traffic surveillance images with highly reflections. In the night, reflections on the road (water) surface, vehicles bodies, or some reflective objects (such as lane markings, traffic signs) will interfere the headlight detection seriously. Although the existing methods have achieved good results, however, most of them failed to detect the headlight when headlights are far from camera. In order to solve the issue, we propose a novel method for vehicle headlights detection. The proposed method makes full use of the brightness and gradient information of the headlights in the night. First, we propose an effective region-of-interest (ROI) segmentation method which is based on multi-scale local saliency detection. The method pre-serve faint or small-sized objects and retain the original shape of the object to the greatest extent. Then, we compute the pyramid histogram of oriented gradients (PHOG) features, which are used to train support vector machine (SVM) classifier. Finally, the extracted bright blocks are classified according to the pre-trained SVM classifier. Experimental results and quantitative evaluations in different scenes demonstrate that our proposed method can achieve a better result compared with previous methods.

Colonoscopic Polyp Classification Using Local Shape and Texture Features

In this paper, a method is proposed for colonic polyp classification which can perform a virtual biopsy for assessing the stage of malignancy in polyps. Geometry, texture, and colour of a polyp give sufficient cue of its nature. The proposed framework characterizes geometry or shape of a polyp by pyramid histogram of oriented gradient (PHOG) features. To encapsulate the texture of the polyp surface, a fractal weighted local binary pattern (FWLBP) descriptor is employed, which is robust to affine transformation. It is also partially robust to illumination variations which is generally encountered during endoscopy. The optimal feature fusion is done using a feature ranking algorithm based on fuzzy entropy. Finally, to evaluate the classification performance of the proposed model, kernel-based support vector machines (SVM) and RUSBoosted tree are used. Experimental results carried on two databases clearly indicate that the proposed method can be used in the colonoscopic polyps classification. The proposed method can give polyp classification accuracies of 90.12% and 84.1%, and AUC of 0.91 and 0.92 for publicly available database and our own database, respectively.

Deep Learning-based Classification of Fruit Diseases: An Application for Precision Agriculture

Agriculture is essential for the economy and plant disease must be minimized. Early recognition of problems is important, but the manual inspection is slow, error-prone, and has high manpower and time requirements. Artificial intelligence can be used to extract fruit color, shape, or texture data, thus aiding the detection of infections. Recently, the convolutional neural network (CNN) techniques show a massive success for image classification tasks. CNN extracts more detailed features and can work efficiently with large datasets. In this work, we used a combined deep neural network and contour feature-based approach to classify fruits and their diseases. A fine-tuned, pretrained deep learning model (VGG19) was retrained using a plant dataset, from which useful features were extracted. Next, contour features were extracted using pyramid histogram of oriented gradient (PHOG) and combined with the deep features using serial based approach. During the fusion process, a few pieces of redundant information were added in the form of features. Then, a “relevance-based” optimization technique was used to select the best features from the fused vector for the final classifications. With the use of multiple classifiers, an accuracy of up to 99.6% was achieved on the proposed method, which is superior to previous techniques. Moreover, our approach is useful for 5G technology, cloud computing, and the Internet of Things (IoT).

Zone-based keyword spotting in Bangla and Devanagari documents

In this paper, we present a word spotting system in text lines for offline Indic scripts such as Bangla (Bengali) and Devanagari. Recently, it was shown that the zone-wise recognition method improves word recognition performance than the conventional full word recognition system in Indic scripts, like Bangla, Devanagari, Gurumukhi (Roy et al. in Pattern Recogn 60: 1057-1075, 26; Bhunia et al. in Pattern Recogn 79: 12–31, 6). Inspired from this idea we consider the zone segmentation approach and use middle zone information to improve the traditional word spotting performance. To avoid the problem of zone segmentation using heuristic approach, we propose here a new HMM based approach to segment the upper and lower zone components from the text line images. The candidate keywords are searched from a line without segmenting characters or words. Also, we propose a feature combining foreground and background information of text line images for keyword-spotting by character filler models. A significant improvement in performance is noted by using both foreground and background information instead of the individual one. Pyramid Histogram of Oriented Gradient (PHOG) feature has been used in our word spotting framework. From the experiment, it has been noted that the proposed zone-segmentation based system outperforms traditional approaches of word spotting.

Mouse Cursor Control Using Hand Gesture Recognition Based on PHOG-Improved LBP and K-NN Classification

A real-time mouse cursor control system using static gestures was designed to achieve the goal of human-computer interaction (HCI) in this paper. The system uses the computer’s own camera or external USB camera to collect video data, detect and recognize gestures of people in the video, and control the cursor movement or click in real time based on the gesture. To improve the recognition accuracy of hand gesture images detection, a method of hand gesture images detection based on PHOG + Improved LBP + K-NN was proposed in this paper. In order to improve the real-time performance of the system, the system determines whether the current frame has human hands by skin color detection. When human hands are detected, Pyramid Histogram of Oriented Gradients (PHOG) features and the improved Local Binary Pattern (LBP) features are further extracted. After fusing PHOG features and improved LBP features, k-nearest neighbor classification (K-NN) is used to implement gesture recognition. Six different gestures were tested 50 times with different angles, different lights and no skin tone in the background. The experimental results show that the system has good recognition performance.

Comparative Study on Deep Convolution Neural Networks DCNN-Based Offline Arabic Handwriting Recognition

Recently, deep learning techniques demonstrated efficiency in building better performing machine learning models which are required in the field of offline Arabic handwriting recognition. Our ancient civilizations presented valuable handwritten manuscripts that need to be documented digitally. If we compared between Latin and the isolated Arabic character recognition, the latter is much more challenging due to the similarity between characters, and the variability of the writing styles. This paper proposes a multi-stage cascading system to serve the field of offline Arabic handwriting recognition. The approach starts with applying the Hierarchical Agglomerative Clustering (HAC) technique to split the database into partially inter-related clusters. The inter-relations between the constructed clusters support representing the database as a big search tree model and help to attain a reduced complexity in matching each test image with a cluster. Cluster members are then ranked based on our new proposed ranking algorithm. This ranking algorithm starts with computing Pyramid Histogram of Oriented Gradients (PHoG), and is followed by measuring divergence by Kullback-Leibler method. Eventually, the classification process is applied only to the highly ranked matching classes. A comparative study is made to assess the effect of six different deep Convolution Neural Networks (DCNNs) on the final recognition rates of the proposed system. Experiments are done using the IFN/ENIT Arabic database. The proposed clustering and ranking stages lead to using only 11% of the whole database in classifying test images. Accordingly, more reduced computation complexity and more enhanced classification results are achieved compared to recent existing systems.