Reduced Feature Vector Research Articles

A hybrid cascade method for EEG classification

The classification of EEG signals is an essential step in the design of brain–computer interface. In order to optimize this step, a method for EEG classification using a cascade technique is proposed. This classification process is implemented in two steps. In the first stage, the feature extraction is accomplished using a discrete wavelet transform (DWT), followed by a feature selection via a filter method in order to select pertinent features from the original feature set. Therefore, reduced feature vectors are presented as input to the classifier. In the second step, features are extracted from the misclassified trials resulting from the first step. Then, the most relevant features are selected. Eventually, the feature vectors are classified. Two configurations of cascade classifiers are tested. In each of them, the classifier used in the first step of the classification scheme is different from the one applied in the second step. Moreover, regarding feature extraction, two cases are studied. In the first case, the same wavelet function is applied in the two steps of the classification process, whereas in the second case, two different wavelet functions are applied successively in the two steps of the classification procedure. The proposed classification scheme was applied on EEG signals of motor imagery dataset III available in BCI competition II, leading to good classification results compared to the existent methods.

Use of Logistic Regression to Determine Parameter Ability for Recovery Prediction in Experimental Rat Hemorrhagic Shock Model

IntroductionHemorrhagic shock (HS) severely decreases oxygen (O2) delivery and induces cardiovascular collapse. It is of interest for clinicians to estimate the potential of recovery during the early onset of a shock event as that would determine whether resources should be expended in environments of material shortage such as warzones and low resource communities. Multiple parameters, such as mean arterial pressure (MAP) and heart rate, are traditionally monitored during shock as a means of estimating the potential for recovery. In this study we used a logistic regression model to determine the ability of systemic hemodynamics, cardiac function, and blood gas parameters in determining the ability of successful HS recovery.MethodsPacked red blood cells (pRBCs) stored in AS‐3 after leukorfiltration were created from donor Sprague‐Dawley rats. pRBC units were randomly stored under either 1) conventional; 2) anaerobic; or 3) anaerobic/hypercapnic conditions. Rats (150–200g) were hemorrhaged to 50% of blood volume, held in hypovolemia for 30 minutes, and resuscitated to recover blood pressure to 90% pre‐hemorrhage with pRBC stored for either 1 or 3 weeks. Systemic hemodynamic, cardiac function, and blood gas parameters were monitored during shock and resuscitation. The percent resuscitation (%Res) corresponding to the percentage of the animal's original blood volume infused at the end of recovery, was used to create two classes of subjects with %Res=50% (equivalent to approximately 5 blood units) set as the threshold. A binary classification, logistic regression model was created with all measured parameters as features, and trained on random instances. The weights with an above threshold contribution assigned to each parameter after an L2 norm cost function minimization were extracted and matched with the corresponding features creating a reduced feature vector. A Monte Carlo approach was applied for n=100 trials of the previously described procedure to get a distribution of the most commonly selected features. The performance of the new feature vector was assessed by another logistic regression model with a 5‐ fold cross‐validation.ResultsThe top 3 features chosen by the model in order of increasing number of occurrences were: lactate 30 minutes into shock, mean arterial pressure 30 minutes into shock and contractility (dp/dtmax) at baseline. The model trained using these three features had a mean cross‐validation accuracy of 90.2%, significantly above the 66.9% accuracy (p<0.001) for a baseline classifier trained with random features. A final test set accuracy of 88.5% was obtained.ConclusionsIt has been shown that a logistic regression model can be effectively used for the determination of important parameters for the prediction of successful recovery after shock in an experimental rat hemorrhagic shock model.Support or Funding InformationNIH: R01 HL126945, NIH: T32 HL105373This abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.

Face recognition algorithm based on wavelet transform and local linear embedding

In this paper, a face recognition fusion algorithm, namely WT-LLE-LSSVM, based on wavelet transform (WT) and local linear embedding (LLE) is proposed. Firstly, the face image is pre-processed, and decomposed by Wavelet Transform to obtain four components of face image; then, the LLE algorithm is carried out to extract the features from the four components respectively, and the weighted fusion is used to obtain the face recognition feature vector; Finally, least squares support vector machine (LSSVM) is conducted to learn the training samples with feature vector reduction, and the classifier of face recognition is established. In the case of different pixels, three face databases are chosen to test some main performance indexes such as recognition rate, speed and time of the algorithm. The simulation results show that compared with the traditional face recognition algorithm, the proposed algorithm can improve the face recognition rate, as well as the computing rate, and enhance the efficiency of face recognition.

DCA‐based unimodal feature‐level fusion of orthogonal moments for Indian sign language dataset

Sign language recognition system classifies signs made by hand gestures. An adequate number of features are required to represent the shape variations of sign language. As compared to individual feature set, a combination of features can be effective due to the fact that a particular feature set represents different shape information. A simple concatenation results in large feature vector size and increases the classification computational complexity. Discriminant correlation analysis (DCA)‐based unimodal feature‐level fusion has been applied on uniform as well as complex background Indian sign language datasets. DCA is a feature‐level fusion technique that takes into account the class associations while combining the feature sets. It maximises the inter‐class separability of two feature sets and also minimises the intra‐class separability while performing the feature fusion. The objective of DCA‐based unimodal feature fusion technique is to combine different feature sets into a single feature vector with more discriminative power. The performance of proposed framework is compared with individual orthogonal moment‐based feature sets and canonical correlation analysis (CCA)‐based feature fusion technique. Results show that in comparison to individual features and CCA‐based fused features, DCA is an effective technique in terms of improved accuracy, reduced feature vector size and smaller classification time.

Comparison of feature extraction techniques for classification of hardwood species

The texture of an image plays an important role in identification and classification of images. The hardwood species of an image contains four key elements namely: vessels (popularly known as pores in cross-section view), fibres, parenchyma's and rays, useful in its identification and classification. Further, the arrangements of all these elements posses texture rich features. Thus, in this work investigation of existing texture feature extraction techniques for the classification of hardwood species have been done. The texture features are extracted from greyscale images of hardwood species to reduce the computational complexity. Further, linear support vector machine (SVM), radial basis function (RBF) kernel SVM, random forest (RF) and linear discriminant analysis (LDA) have been employed as classifiers to investigate the efficacy of the texture feature extraction techniques. The classification accuracy of the existing texture descriptors has been compared. Further, principal component analysis (PCA) and minimal-redundancy-maximal-relevance (mRMR) feature selection method is employed to select the best subset of feature vector data. The PCA reduced feature vector data of co-occurrence of adjacent local binary pattern (CoALBP24) texture feature extraction technique has attained maximum classification accuracy of 96.33 ± 1.14% with the help of LDA classifier.

A novel approach for dimension reduction of microarray

This paper proposes a new hybrid search technique for feature (gene) selection (FS) using Independent component analysis (ICA) and Artificial Bee Colony (ABC) called ICA+ABC, to select informative genes based on a Naïve Bayes (NB) algorithm. An important trait of this technique is the optimization of ICA feature vector using ABC. ICA+ABC is a hybrid search algorithm that combines the benefits of extraction approach, to reduce the size of data and wrapper approach, to optimize the reduced feature vectors. This hybrid search technique is facilitated by evaluating the performance of ICA+ABC on six standard gene expression datasets of classification. Extensive experiments were conducted to compare the performance of ICA+ABC with the results obtained from recently published Minimum Redundancy Maximum Relevance (mRMR) +ABC algorithm for NB classifier. Also to check the performance that how ICA+ABC works as feature selection with NB classifier, compared the combination of ICA with popular filter techniques and with other similar bio inspired algorithm such as Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). The result shows that ICA+ABC has a significant ability to generate small subsets of genes from the ICA feature vector, that significantly improve the classification accuracy of NB classifier compared to other previously suggested methods.

A comparison of soft computing models for Parkinson’s disease diagnosis using voice and gait features

Parkinson’s disease is a widespread disease among elder population worldwide caused by dopamine loss, which reduces quality of life because of motor and non-motor complications. In the current paper, nine soft computing models, i.e., Cubist, Cubist Committees, Random Forests, Kernel Support Vector Machine, Linear Regression, Naive Bayes, Artificial Neural Network, Adaptive Neuro-Fuzzy Inference System, and Hybrid Neuro-Fuzzy Inference System are implemented for Parkinson’s disease diagnosis using voice and gait features. Later, their performances are evaluated based on performance measures, viz., true positive, false positive, false negative, true negative, accuracy, sensitivity, specificity, and RMSE, and finally, a comparison is performed to identify the most efficient model and data set combination. The comparison demonstrated that Random Forest model outperformed others yielding 100% accuracy, 100% sensitivity, 100% specificity, and zero RMSE on voice and gait training data sets both; Cubist Committees model outperformed others yielding 74.00% accuracy, 69.39% sensitivity, 78.43% specificity, and 0.4582 RMSE on voice testing data set; Random Forest model once again outperformed others yielding 81.66% accuracy, 92.39% sensitivity, 66.67% specificity, and 0.4283 RMSE on gait testing data set. Furthermore, these models’ performances are also evaluated on reduced feature vector voice and gait data sets obtained by Principal Component Analysis, and compared with their performances on former data sets. The comparison exhibited that the soft computing models’ performance decreases by reducing feature vector of the data sets.

Odia Handwritten Vowel Recognition System using Slantlet Transform and Differential Evolution based Functional Link Artificial Neural Network Classifier

One of the challenging problems in optical character recognition (OCR) is the identification of handwritten characters. It has several complexities like missing part, size variation, added noise, media devices used, slant variation etc. The goal of this paper is to develop an effective system for recognition of Odia handwritten vowels using functional link artificial neural network (FLANN) classifier and differential evolution (DE) optimization technique. In this paper the emphasis is on preprocessing, feature extraction, feature reduction and classification with optimization technique. Various preprocessing operations like size normalization mean filtering and canny edge detection methods are carried out on the vowel images to enhance the quality. Slantlet transform (SLT) is applied for feature extraction and the extracted features are further reduced by principal component analysis (PCA) method. FLANN classifier is applied on the reduced feature vector and the weights of the FLANN are optimized with DE evolutionary technique. The system has achieved a good classification accuracy of 91.32% on the dataset.

Domain invariant feature extraction against evasion attack

In the security application, an attacker might violate the data stationary assumption that is a common assumption in the most machine learning techniques. This problem named as the domain shift problem arises when training (source) and test (target) data follow different distributions. The inherent adversarial nature of the security applications considerably effects on the robustness of a learning system. For that, a classifier designer needs to evaluate the robustness of a learning system under potential attacks during the design phase. The previous studies investigate the effect of reduced feature vector on the security evaluation of a learning classifier. They demonstrate that traditional feature selection techniques lead to even worsen performance. Therefore, an adversary-aware feature selection algorithm is proposed to improve the robustness of the learning systems. However, prior studies in domain adaptation techniques which are fundamental in addressing domain shift problem demonstrate that original space may not be directly suitable for refining this distribution mismatch, because some features may have been distorted by the domain shift. In this paper, we propose domain invariant feature extraction model based on domain adaptation technique in order to address domain shift problem caused by an adversary. We conduct an experiment that graphically shows the effect of a successful attack on the MNIST handwritten digits classification task. After that, we design synthetic datasets to investigate the effect of reduced feature vector on the performance of a learning system under attack. Moreover, our proposed feature extraction model significantly outperforms the adversarial-aware feature selection and traditional feature selection models on the application of spam filtering

Parkinsons Disease Diagnosis by Adaptive Boosting and Classification Tree using Voice Features

Parkinson’s disease is a widespread disease among elder population worldwide effecting approximately 6.3 million people across all genders, races and cultures. It is caused by dopamine loss, a chemical mediator that is responsible for body’s ability to control the movements. The disease reduces quality of life because of motor and non-motor complications. In this article Adaptive Boosting and Classification Tree based soft computing models are implemented to diagnose Parkinson’s disease using voice features. The soft computing models performances are evaluated on performance measures viz. true positive, false positive, false negative, true negative, accuracy, sensitivity, specificity, RMSE on training and datasets. Finally a comparison is performed to identify the most efficient model and dataset combination. Adaptive Boosting model outperformed others on reduced feature vector dataset obtained by selecting prominent 15 principal components using principal component analysis, where, it demonstrated 100% accuracy, 100% sensitivity, 100% specificity, 0.0 RMSE on training dataset and 67.00% accuracy, 67.35% sensitivity, 66.67% specificity, 0.5745 RMSE on testing dataset.

Fault diagnosis for rolling bearing based on SIFT-KPCA and SVM

PurposeThe purpose of this paper is to propose a fault diagnosis method for rolling bearings, in which the fault feature extraction is realized in a two-dimensional domain using scale invariant feature transform (SIFT) algorithm. This method is different from those methods extracting fault feature directly from the traditional one-dimensional domain.Design/methodology/approachThe vibration signal of rolling bearings is first transformed into a two-dimensional image. Then, the SIFT algorithm is applied to the image to extract the scale invariant feature vector which is highly distinctive and insensitive to noises and working condition variation. As the extracted feature vector is high-dimensional, kernel principal component analysis (KPCA) algorithm is utilized to reduce the dimension of the feature vector, and singular value decomposition technique is used to extract the singular values of the reduced feature vector. Finally, these singular values are introduced into a support vector machine (SVM) classifier to realize fault classification.FindingsThe experiment results show a high fault classification accuracy based on the proposed method.Originality/valueThe proposed approach for rolling bearing fault diagnosis based on SIFT-KPCA and SVM is highly effective in the experiment. The practical value in engineering application of this method can be researched in the future.

Subject-based discriminative sparse representation model for detection of concealed information

Background and objectivesThe use of machine learning approaches in concealed information test (CIT) plays a key role in the progress of this neurophysiological field. In this paper, we presented a new machine learning method for CIT in which each subject is considered independent of the others. The main goal of this study is to adapt the discriminative sparse models to be applicable for subject-based concealed information test. MethodsIn order to provide sufficient discriminability between guilty and innocent subjects, we introduced a novel discriminative sparse representation model and its appropriate learning methods. For evaluation of the method forty-four subjects participated in a mock crime scenario and their EEG data were recorded. As the model input, in this study the recurrence plot features were extracted from single trial data of different stimuli. Then the extracted feature vectors were reduced using statistical dependency method. The reduced feature vector went through the proposed subject-based sparse model in which the discrimination power of sparse code and reconstruction error were applied simultaneously. ResultsExperimental results showed that the proposed approach achieved better performance than other competing discriminative sparse models. The classification accuracy, sensitivity and specificity of the presented sparsity-based method were about 93%, 91% and 95% respectively. ConclusionsUsing the EEG data of a single subject in response to different stimuli types and with the aid of the proposed discriminative sparse representation model, one can distinguish guilty subjects from innocent ones. Indeed, this property eliminates the necessity of several subject EEG data in model learning and decision making for a specific subject.

Heartbeat classification system based on neural networks and dimensionality reduction

Introduction This paper presents a complete approach for the automatic classification of heartbeats to assist experts in the diagnosis of typical arrhythmias, such as right bundle branch block, left bundle branch block, premature ventricular beats, premature atrial beats and paced beats. Methods A pre-processing step was performed on the electrocardiograms (ECG) for baseline removal. Next, a QRS complex detection algorithm was implemented to detect the heartbeats, which contain the primary information that is employed in the classification approach. Next, ECG segmentation was performed, by which a set of features based on the RR interval and the beat waveform morphology were extracted from the ECG signal. The size of the feature vector was reduced by principal component analysis. Finally, the reduced feature vector was employed as the input to an artificial neural network. Results Our approach was tested on the Massachusetts Institute of Technology arrhythmia database. The classification performance on a test set of 18 ECG records of 30 min each achieved an accuracy of 96.97%, a sensitivity of 95.05%, a specificity of 90.88%, a positive predictive value of 95.11%, and a negative predictive value of 92.7%. Conclusion The proposed approach achieved high accuracy for classifying ECG heartbeats and could be used to assist cardiologists in telecardiology services. The main contribution of our classification strategy is in the feature selection step, which reduced classification complexity without major changes in the performance.

Diagnosis of Alzheimer's Disease Using Dual-Tree Complex Wavelet Transform, PCA, and Feed-Forward Neural Network.

Background. Error-free diagnosis of Alzheimer’s disease (AD) from healthy control (HC) patients at an early stage of the disease is a major concern, because information about the condition’s severity and developmental risks present allows AD sufferer to take precautionary measures before irreversible brain damage occurs. Recently, there has been great interest in computer-aided diagnosis in magnetic resonance image (MRI) classification. However, distinguishing between Alzheimer’s brain data and healthy brain data in older adults (age > 60) is challenging because of their highly similar brain patterns and image intensities. Recently, cutting-edge feature extraction technologies have found extensive application in numerous fields, including medical image analysis. Here, we propose a dual-tree complex wavelet transform (DTCWT) for extracting features from an image. The dimensionality of feature vector is reduced by using principal component analysis (PCA). The reduced feature vector is sent to feed-forward neural network (FNN) to distinguish AD and HC from the input MR images. These proposed and implemented pipelines, which demonstrate improvements in classification output when compared to that of recent studies, resulted in high and reproducible accuracy rates of 90.06 ± 0.01% with a sensitivity of 92.00 ± 0.04%, a specificity of 87.78 ± 0.04%, and a precision of 89.6 ± 0.03% with 10-fold cross-validation.

Influence of feature set reduction on breast cancer malignancy classification of fine needle aspiration biopsies

Grading of breast cancer malignancy is a key step in its diagnosis, which in turn helps to determine its prognosis and a course of treatment. In this paper, we consider the application of pattern recognition and image processing techniques to perform computer-assisted automatic breast cancer malignancy grading from cytological slides of fine needle aspiration biopsies. To determine a classification of the malignancy of the slide, a feature set is first determined from imagery of the slides. In this paper we investigated the nature of a wide set of features extracted from biopsy images to determine their discriminatory power and cross-correlation. Feature vector reduction is studied using a correlation map of the features, determining discriminatory power using the Kolmogorov–Smirnov test, significant feature selection, and stepwise feature selection. The reduction of the feature vector simplifies the complexity of classification scheme and does not impair the classification accuracy. In some cases a decrease of the error rate is noted. Based on this analysis, we present an improved classification system for cancer malignancy grading.

Image driven machine learning methods for microstructure recognition

Computer vision and machine learning methods were applied to the challenge of automatic microstructure recognition. Here, a case study on dendritic morphologies was performed. Two classification tasks were completed, and involved distinguishing between micrographs that depict dendritic morphologies from those that do not contain this particular microstructural feature (Task 1), and from those micrographs identified as depicting dendrites, different cross-sectional views (longitudinal or transverse) were identified (Task 2). Data sets were comprised of images taken over a range of magnifications, from materials with different compositions and varying orientations of microstructural features. Feature extraction and dimensionality reduction were performed prior to training machine learning algorithms to classify microstructural image data. Visual bag of words, texture and shape statistics, and pre-trained convolutional neural networks (deep learning algorithms) were used for feature extraction. Classification was then performed using support vector machine, voting, nearest neighbors, and random forest models. For each model, classification was completed using full (original size) and reduced feature vectors for each feature extraction method tested. Performance comparisons were done to evaluate all possible combinations of feature extraction, selection, and classifiers for the task of micrograph classification. Results demonstrate that pre-trained neural networks represent microstructure image data well, and when used for feature extraction yield the highest classification accuracies for the majority of classifier and feature selection methods tested. Thus, deep learning algorithms can successfully be applied to micrograph recognition tasks. Maximum classification accuracies of 91.85±4.25% and 97.37±3.33% for Tasks 1 and 2 respectively, were achieved. This work is a broad investigation of computer vision and machine learning methods that acts as a step towards applying these established methods to more sophisticated materials recognition or characterization tasks. The approach presented here could offer improvements over established stereological measurements by removing the requirement of expert knowledge (bias) for interpretation of image data prior to characterization.

Systemic health evaluation of RF generators using Gaussian mixture models

We propose an application of specific machine learning techniques capable of evaluating systemic health of a Radio Frequency (RF) power generator. System signatures or fingerprints are collected from multivariate time-series data samples of sensor values under typical operational loads. These fingerprints are transformed into feature vectors using standard scaling/translation methods and the Fast Fourier Transform (FFT). The number of features per fingerprint are reduced by banding neighboring features and Principal Component Analysis (PCA). The reduced feature vectors are used with the Expectation Maximization (EM) algorithm to learn parameters for a Gaussian Mixture Model (GMM) to represent normal operation. One-class classification of normal fingerprints is achieved by thresholding the likelihood of a fingerprint feature vectors. Fingerprints were collected from normal operational conditions and seeded non-normal conditions. Preprocessing methods and algorithmic parameters have been selected using an iterative grid search. Average robust true positive rate achieved was 94.76% and best specificity reported is 86.56%.

An Improved Approach for Text-Independent Speaker Recognition

This paper presents new Speaker Identification and Speaker Verification systems based on the use of new feature vectors extracted from the speech signal. The proposed structure combine between the most successful Mel Frequency Cepstral Coefficients and new features which are the Short Time Zero Crossing Rate of the signal. A comparison between speaker recognition systems based on Gaussian mixture models using the well known Mel Frequency Cepstral Coefficients and the novel systems based on the use of a combination between both reduced Mel Frequency Cepstral Coefficients features vectors and Short Time Zero Crossing Rate features is given. This comparison proves that the use of the new reduced feature vectors help to improve the system’s performance and also help to reduce the time and memory complexity of the system which is required for realistic applications that suffer from computational resource limitation. The experiments were performed on speakers from TIMIT database for different training durations. The suggested systems performances are evaluated against the baseline systems. The increase of the proposed systems performances are well observed for identification experiments and the decrease of Equal Error Rates are also remarkable for verification experiments. Experimental results demonstrate the effectiveness of the new approach which avoids the use of more complex algorithms or the combination of different approaches requiring lengthy calculation.

An Automated and Intelligent Medical Decision Support System for Brain MRI Scans Classification.

A wide interest has been observed in the medical health care applications that interpret neuroimaging scans by machine learning systems. This research proposes an intelligent, automatic, accurate, and robust classification technique to classify the human brain magnetic resonance image (MRI) as normal or abnormal, to cater down the human error during identifying the diseases in brain MRIs. In this study, fast discrete wavelet transform (DWT), principal component analysis (PCA), and least squares support vector machine (LS-SVM) are used as basic components. Firstly, fast DWT is employed to extract the salient features of brain MRI, followed by PCA, which reduces the dimensions of the features. These reduced feature vectors also shrink the memory storage consumption by 99.5%. At last, an advanced classification technique based on LS-SVM is applied to brain MR image classification using reduced features. For improving the efficiency, LS-SVM is used with non-linear radial basis function (RBF) kernel. The proposed algorithm intelligently determines the optimized values of the hyper-parameters of the RBF kernel and also applied k-fold stratified cross validation to enhance the generalization of the system. The method was tested by 340 patients’ benchmark datasets of T1-weighted and T2-weighted scans. From the analysis of experimental results and performance comparisons, it is observed that the proposed medical decision support system outperformed all other modern classifiers and achieves 100% accuracy rate (specificity/sensitivity 100%/100%). Furthermore, in terms of computation time, the proposed technique is significantly faster than the recent well-known methods, and it improves the efficiency by 71%, 3%, and 4% on feature extraction stage, feature reduction stage, and classification stage, respectively. These results indicate that the proposed well-trained machine learning system has the potential to make accurate predictions about brain abnormalities from the individual subjects, therefore, it can be used as a significant tool in clinical practice.

Comprehensive Performance Comparison of Fourier, Walsh, Haar, Sine and Cosine Transforms for Video Retrieval with Partial Coefficients of Transformed Video

desire of better and faster retrieval techniques has always fuelled to the research in content based video retrieval (CBVR). The extended comparison of innovative content based video retrieval (CBVR) techniques based on feature vectors as partial coefficients of transformed video frames using various orthogonal transforms is presented in the paper. Here the popular transforms are considered like Cosine, Walsh, Haar, Sine, and Fourier transforms. The advantage of energy compaction of transforms in higher energy coefficients is taken to reduce the feature vector size per video by taking partial coefficients of transformed video frames. Reduced feature vector size results in less time for comparison of feature vectors resulting in faster retrieval of videos. The features are extracted in eight different ways from the transformed image. First all the coefficients of transformed image considered as 100% energy and then seven reduced coefficients sets are considered as feature vectors (as 99%, 98%, 97%, 96%, 95%, 90% and 85% energy of complete transformed video coefficients). To extract Gray feature sets the five video transforms are applied on gray image equivalents and the color components of videos. Then these seven reduced coefficients sets are used instead of using all coefficients of transformed videos as feature vector for video retrieval, resulting into better performance and lower computations. The video database of 500 video spread across 10 categories is used to test the performance of proposed CBVR techniques. 500 queries are fired on the database to find average accuracy values for all feature sets per transform for each proposed CBVR technique. The results have shown performance improvement (higher accuracy values) with partial coefficients compared to complete 100% energy of transformed of video frames at reduced computations resulting in faster retrieval. Haar transform surpasses all other considered transforms in performance with highest accuracy values with 90% of partial energy coefficients and size is lowered by 99.93% as compared to other transforms.