Features For Image Classification Research Articles

Parallel and Efficient Sensitivity Analysis of Microscopy Image Segmentation Workflows in Hybrid Systems.

We investigate efficient sensitivity analysis (SA) of algorithms that segment and classify image features in a large dataset of high-resolution images. Algorithm SA is the process of evaluating variations of methods and parameter values to quantify differences in the output. A SA can be very compute demanding because it requires re-processing the input dataset several times with different parameters to assess variations in output. In this work, we introduce strategies to efficiently speed up SA via runtime optimizations targeting distributed hybrid systems and reuse of computations from runs with different parameters. We evaluate our approach using a cancer image analysis workflow on a hybrid cluster with 256 nodes, each with an Intel Phi and a dual socket CPU. The SA attained a parallel efficiency of over 90% on 256 nodes. The cooperative execution using the CPUs and the Phi available in each node with smart task assignment strategies resulted in an additional speedup of about 2×. Finally, multi-level computation reuse lead to an additional speedup of up to 2.46× on the parallel version. The level of performance attained with the proposed optimizations will allow the use of SA in large-scale studies.

Correlating and Modeling of Extracted Features from PVT Images of Composites using Optical Flow Technique and Weight Elimination Algorithm Optimization [OFT-WEA

A new approach to the use and implementation of Optical Flow technique is presented. The technique extracts features from presented images as a function of reference image and produces percentage of matching between the reference and tested images. The new approach in using Optical Flow lies in replacing the motion part of the algorithm with differential time related changes in an infrared thermal image sequence with frames of images taken as a result of applying the Pulse Video Thermography (PVT) technique. The sequence of images or frames is obtained for the tested structures of composites before and after impact damage. The resulted data of the tested images is used to establish mathematical model that can be used to predict impact energy from collected features or predict expected features from knowing impact damage level. To optimize the mathematical model, a new way of using Neural Networks is employed, which aims at obtaining a best fit for the used variables in the mathematical model, hence resulting in a better testing interpretation and more accurate prediction and classification of image features to improve future composite structures designs. The Neural Network Weigh Elimination Algorithm (WEA) is used and proved effective in predicting areas of damage.

Colour face recognition using fuzzy quaternion-based discriminant analysis

Colour information has been shown to be effective in improving object recognition performance. In this paper, we propose a novel quaternion-based colour model with enhanced fuzzy parameterized discriminant analysis to perform face recognition. The proposed method represents and classifies colour images by using an improved fuzzy quaternion-based discriminant (FQD) model, which is effective for colour image feature representation, extraction and classification. More specifically, each pixel in a colour image is first assigned a quaternion number, and a quaternion-based vector is then generated to represent this colour image. Second, an enhanced fuzzy parameterized discriminant analysis is used to transform the original quaternion-based vector into an optimized discriminant quaternion space. Third, colour face recognition is conducted by interpreting the colour feature model as fuzzy weight measurement in a quaternion discriminant analysis. The main contribution of this paper is that it provides a novel fuzzy supervised learning approach to reconstruct the quaternion-based discriminant vector space, thus showing the importance of the FQD characteristic from colour spaces for colour-image-based face recognition. Experimental results on the AR and Georgia Tech colour datasets demonstrate the effectiveness of the proposed method.

Feature analysis for classification of trace fluorescent labeled protein crystallization images

BackgroundLarge number of features are extracted from protein crystallization trial images to improve the accuracy of classifiers for predicting the presence of crystals or phases of the crystallization process. The excessive number of features and computationally intensive image processing methods to extract these features make utilization of automated classification tools on stand-alone computing systems inconvenient due to the required time to complete the classification tasks. Combinations of image feature sets, feature reduction and classification techniques for crystallization images benefiting from trace fluorescence labeling are investigated.ResultsFeatures are categorized into intensity, graph, histogram, texture, shape adaptive, and region features (using binarized images generated by Otsu’s, green percentile, and morphological thresholding). The effects of normalization, feature reduction with principle components analysis (PCA), and feature selection using random forest classifier are also analyzed. The time required to extract feature categories is computed and an estimated time of extraction is provided for feature category combinations. We have conducted around 8624 experiments (different combinations of feature categories, binarization methods, feature reduction/selection, normalization, and crystal categories). The best experimental results are obtained using combinations of intensity features, region features using Otsu’s thresholding, region features using green percentile G90 thresholding, region features using green percentile G99 thresholding, graph features, and histogram features. Using this feature set combination, 96% accuracy (without misclassifying crystals as non-crystals) was achieved for the first level of classification to determine presence of crystals. Since missing a crystal is not desired, our algorithm is adjusted to achieve a high sensitivity rate. In the second level classification, 74.2% accuracy for (5-class) crystal sub-category classification. Best classification rates were achieved using random forest classifier.ContributionsThe feature extraction and classification could be completed in about 2 s per image on a stand-alone computing system, which is suitable for real time analysis. These results enable research groups to select features according to their hardware setups for real-time analysis.

Segmentation and Classification of Retinal Image Features

Segmentation and Classification of Retinal Image Features

Gender Recognition from Human-Body Images Using Visible-Light and Thermal Camera Videos Based on a Convolutional Neural Network for Image Feature Extraction.

Extracting powerful image features plays an important role in computer vision systems. Many methods have previously been proposed to extract image features for various computer vision applications, such as the scale-invariant feature transform (SIFT), speed-up robust feature (SURF), local binary patterns (LBP), histogram of oriented gradients (HOG), and weighted HOG. Recently, the convolutional neural network (CNN) method for image feature extraction and classification in computer vision has been used in various applications. In this research, we propose a new gender recognition method for recognizing males and females in observation scenes of surveillance systems based on feature extraction from visible-light and thermal camera videos through CNN. Experimental results confirm the superiority of our proposed method over state-of-the-art recognition methods for the gender recognition problem using human body images.

Image analysis using new set of separable two-dimensional discrete orthogonal moments based on Racah polynomials

In this paper, we propose three new separable two-dimensional discrete orthogonal moments baptized: RTM (Racah-Tchebichef moments), RKM (Racah-Krawtchouk moments), and RdHM (Racah-dual Hahn moments). We present a comparative study between our proposed separable two-dimensional discrete orthogonal moments and the classical ones, in terms of gray-level image reconstruction accuracy, including noisy and noise-free conditions. Furthermore, in this study, the local feature extraction capabilities of the proposed moments are described. Finally, a new set of RST (rotation, scaling, and translation) invariants, based on separable proposed moments, is introduced in this paper for the first time, and their description performances are highly tested as pattern features for image classification in comparison with the traditional moment invariants. The experimental results show that the new set of moments is potentially useful in the field of image analysis.

Medical Image Feature Classification from Rule Mining and Retrieval via Weighted Swarm Neural Network

Medical Image Feature Classification from Rule Mining and Retrieval via Weighted Swarm Neural Network

A quantitative symmetry‐based analysis of hyperacute ischemic stroke lesions in noncontrast computed tomography

Early identification of ischemic stroke plays a significant role in treatment and potential recovery of damaged brain tissue. In noncontrast CT (ncCT), the differences between ischemic changes and healthy tissue are usually very subtle during the hyperacute phase (< 8 h from the stroke onset). Therefore, visual comparison of both hemispheres is an important step in clinical assessment. A quantitative symmetry-based analysis of texture features of ischemic lesions in noncontrast CT images may provide an important information for differentiation of ischemic and healthy brain tissue in this phase. One hundred thirty-nine (139) ncCT scans of hyperacute ischemic stroke with follow-up magnetic resonance diffusion-weighted (MR-DW) images were collected. The regions of stroke were identified in the MR-DW images, which were spatially aligned to corresponding ncCT images. A state-of-the-art symmetric diffeomorphic image registration was utilized for the alignment of CT and MR-DW, for identification of individual brain hemispheres, and for localization of the region representing healthy tissue contralateral to the stroke cores. Texture analysis included extraction and classification of co-occurrence and run-length texture-based image features in the regions of ischemic stroke and their contralateral regions. The classification schemes achieved area under the receiver operating characteristic [Az] ≈ 0.82 for the whole dataset. There was no statistically significant difference in the performance of classifiers for the data sets with time between 2 and 8 hours from symptom onset. The performance of the classifiers did not depend on the size of the stroke regions. The results provide a set of optimal texture features which are suitable for distinguishing between hyperacute ischemic lesions and their corresponding contralateral brain tissue in noncontrast CT. This work is an initial step toward development of an automated decision support system for detection of hyperacute ischemic stroke lesions on noncontrast CT of the brain.

Estimating False Positive Contamination in Crater Annotations from Citizen Science Data

Web-based citizen science often involves the classification of image features by large numbers of minimally trained volunteers, such as the identification of lunar impact craters under the Moon Zoo project. Whilst such approaches facilitate the analysis of large image data sets, the inexperience of users and ambiguity in image content can lead to contamination from false positive identifications. We give an approach, using Linear Poisson Models and image template matching, that can quantify levels of false positive contamination in citizen science Moon Zoo crater annotations. Linear Poisson Models are a form of machine learning which supports predictive error modelling and goodness-of-fits, unlike most alternative machine learning methods. The proposed supervised learning system can reduce the variability in crater counts whilst providing predictive error assessments of estimated quantities of remaining true verses false annotations. In an area of research influenced by human subjectivity, the proposed method provides a level of objectivity through the utilisation of image evidence, guided by candidate crater identifications.

Unsupervised High-Level Feature Extraction of SAR Imagery With Structured Sparsity Priors and Incremental Dictionary Learning

Sparse representation is an effective model for high-level feature extraction, and the dictionary is critical, since it can provide a sparse and discriminative feature for image classification. However, the traditional sparse model with $\ell_{1}$ - norm is unstable and ignores spatial context dependence. Furthermore, the traditional off-line dictionary learning is less efficient. In this letter, a high-level feature extraction approach is proposed, in which structured sparsity priors are imposed on the sparse representation to exploit the context dependence and an incremental structured dictionary learning method is proposed to exploit the inherent structures of a dictionary. The experiment results on unsupervised synthetic aperture radar imagery classification show that the structured priors improve classification performance and the proposed algorithm is more efficient in dictionary learning compared with existing works.

2D Motif Basis Applied to the Classification of Digital Images

The classification of raw data often involves the problem of selecting the appropriate set of features to represent the input data. Different types of features can be extracted from the input dataset, but only some of them are actually relevant for the classification process. Since relevant features are often unknown in real-world problems, many candidate features are usually introduced. This degrades both the speed and the predictive accuracy of the classifier due to the presence of redundancy in the set of candidate features. Recently, a special class of bidimensional motifs, i.e. 2D motif basis has been introduced in the literature. 2D motif basis showed to be powerful in capturing the relevant information of digital images, also achieving good performances for image compression. Here, we investigate the effectiveness of 2D motif basis, when they are used as features for image classification. We embed such features in a bag-of-words model, and then we apply K-Nearest Neighbour for the classification step. Results obtained on both benchmark image datasets and video frames datasets show that, despite the pixel-level nature of the considered features, the achieved accuracy is high and comparable with that of other techniques proposed in the literature.

A combinatorial Bayesian and Dirichlet model for prostate MR image segmentation using probabilistic image features

Blurred boundaries and heterogeneous intensities make accurate prostate MR image segmentation problematic. To improve prostate MR image segmentation we suggest an approach that includes: (a) an image patch division method to partition the prostate into homogeneous segments for feature extraction; (b) an image feature formulation and classification method, using the relevance vector machine, to provide probabilistic prior knowledge for graph energy construction; (c) a graph energy formulation scheme with Bayesian priors and Dirichlet graph energy and (d) a non-iterative graph energy minimization scheme, based on matrix differentiation, to perform the probabilistic pixel membership optimization. The segmentation output was obtained by assigning pixels with foreground and background labels based on derived membership probabilities. We evaluated our approach on the PROMISE-12 dataset with 50 prostate MR image volumes. Our approach achieved a mean dice similarity coefficient (DSC) of 0.90 ± 0.02, which surpassed the five best prior-based methods in the PROMISE-12 segmentation challenge.

Projective robust nonnegative factorization

Nonnegative matrix factorization (NMF) has been successfully used in many fields as a low-dimensional representation method. Projective nonnegative matrix factorization (PNMF) is a variant of NMF that was proposed to learn a subspace for feature extraction. However, both original NMF and PNMF are sensitive to noise and are unsuitable for feature extraction if data is grossly corrupted. In order to improve the robustness of NMF, a framework named Projective Robust Nonnegative Factorization (PRNF) is proposed in this paper for robust image feature extraction and classification. Since learned projections can weaken noise disturbances, PRNF is more suitable for classification and feature extraction. In order to preserve the geometrical structure of original data, PRNF introduces a graph regularization term which encodes geometrical structure. In the PRNF framework, three algorithms are proposed that add a sparsity constraint on the noise matrix based on L1/2 norm, L1 norm, and L2, 1 norm, respectively. Robustness and classification performance of the three proposed algorithms are verified with experiments on four face image databases and results are compared with state-of-the-art robust NMF-based algorithms. Experimental results demonstrate the robustness and effectiveness of the algorithms for image classification and feature extraction.

Tumor Diagnosis Based on the GMM Feature Decision Classification of Brain MR Images

The MR image has provided lots of information used for medical examination. Accurate and robust brain MR image segmentation, feature extraction and feature classification are very important for clinical tumor diagnosis. A new tumor diagnosis method based on brain MR images is hereby put forward. Firstly, detect the deformed area of the images through multi-threshold segmentation morphology, and then, extract the GMM feature used for the classification, and finally, classify the types of tumor images by using decision tree classifier. The whole classification consists of two stages, during training stage extract the different features of tumor images and non-tumor images, and during testing stage conduct the classification of tumor and non-tumor based on knowledge databank. The computing method is appraised according to the three performance index including accuracy, false alarm rate and loss detecting rate, the experiment results show that the computing function is excellent and is helpful for better brain tumor diagnosis.

Brain Tumor Detection and Classification in Magnetic Resonance Imaging (MRI) using Region Growing, Fuzzy Symmetric Measure, and Artificial Neural Network Backpropagation

&lt;p&gt;Brain tumor is one type of malignant tumors that occurs because there is an abnormal and uncontrolled cell division activity. There are several ways to diagnose brain tumors, for example use MRI images. Through the MRI images, the radiologist can see the brain anatomy without performing surgery. However, this process is still done manually and could lead to misdiagnose. In addition, the different characteristics of brain tumor makes the diagnose more difficult. Therefore, we need a system of Computer-Aided Diagnostic (CAD) that will help radiologist in identifying brain tumors. &lt;/p&gt;&lt;p&gt;In general, the CAD system consists of two major processes, namely image segmentation and feature extraction and classification. One example of segmentation is Region Growing that will classify the pixels based on certain criteria. However, the manual selection of seed point is a drawback of this method. The examples of feature extraction methods are Fuzzy Symmetric Measure (FSM), and First and Second Order Statistics. FSM values can be used to calculate the symmetry of the image brain, while the first and second order to represent feature in the image. As for the classification process, Artificial Neural Network Backpropagation method is widely used for its ability to resolve nonlinear dan complex problems.&lt;/p&gt;&lt;p&gt;This research implements CAD system that uses Region Growing, Symmetric Fuzzy Measure, and Backpropagation Neural Network for detecting and classifying the brain tumors. In addition, the modification of converging square is conducted to select a seed point automatically. After testing, the system generates a 100% accuracy and BER is 0 in the case of distinguishing between normal and tumor brain. Besides, the average accuracy in classifying the types of brain tumors achieved 89.72% , the BER 0.1 for training data, and the average accuracy of 84.44%, BER 0.16 for the testing data.&lt;/p&gt;

Metric Embedded Discriminative Vocabulary Learning for High-Level Person Representation

A variety of encoding methods for bag of word (BoW) model have been proposed to encode the local features in image classification. However, most of them are unsupervised and just employ k-means to form the visual vocabulary, thus reducing the discriminative power of the features. In this paper, we propose a metric embedded discriminative vocabulary learning for high-level person representation with application to person re-identification. A new and effective term is introduced which aims at making the same persons closer while different ones farther in the metric space. With the learned vocabulary, we utilize a linear coding method to encode the image-level features (or holistic image features) for extracting high-level person representation. Different from traditional unsupervised approaches, our method can explore the relationship(same or not) among the persons. Since there is an analytic solution to the linear coding, it is easy to obtain the final high-level features. The experimental results on person re-identification demonstrate the effectiveness of our proposed algorithm.

Bag-of-Words and Object-Based Classification for Cloud Extraction From Satellite Imagery

Automatic cloud extraction from satellite imagery is an important task for many applications in remote sensing. Humans can easily identify various clouds from satellite images based on the visual features of cloud. In this study, a method of automatic cloud detection is proposed based on object classification of image features. An image is first segmented into superpixels so that the descriptor of each superpixel can be computed to form a feature vector for classification. The support vector machine algorithm is then applied to discriminate cloud and noncloud regions. Thereafter, the GrabCut algorithm is used to extract more accurate cloud regions. The key of the method is to deal with the highly varying patterns of clouds. The bag-of-words (BOW) model is used to construct the compact feature vectors from densely extracted local features, such as dense scale-invariant feature transform (SIFT). The algorithm is tested using 101 RapidEye and 86 Landsat images with many cloud patterns. These images achieve 89.2% of precision, 87.8% of recall for RapidEye, 85.8% of precision, and 83.9% of recall for Landsat. The experiments show that the method is insensitive to the number of codewords in the codebook construction of the BOW.

Lumbar Ultrasound Image Feature Extraction and Classification with Support Vector Machine

Needle entry site localization remains a challenge for procedures that involve lumbar puncture, for example, epidural anesthesia. To solve the problem, we have developed an image classification algorithm that can automatically identify the bone/interspinous region for ultrasound images obtained from lumbar spine of pregnant patients in the transverse plane. The proposed algorithm consists of feature extraction, feature selection and machine learning procedures. A set of features, including matching values, positions and the appearance of black pixels within pre-defined windows along the midline, were extracted from the ultrasound images using template matching and midline detection methods. A support vector machine was then used to classify the bone images and interspinous images. The support vector machine model was trained with 1,040 images from 26 pregnant subjects and tested on 800 images from a separate set of 20 pregnant patients. A success rate of 95.0% on training set and 93.2% on test set was achieved with the proposed method. The trained support vector machine model was further tested on 46 off-line collected videos, and successfully identified the proper needle insertion site (interspinous region) in 45 of the cases. Therefore, the proposed method is able to process the ultrasound images of lumbar spine in an automatic manner, so as to facilitate the anesthetists' work of identifying the needle entry site.

Multi-level K-means clustering and group sparse coding with quasi-SIFT feature for image classification

Traditional sparse coding treats local descriptors separately, leading to random locations of non-zero coefficients. To overcome this problem, group sparse coding was proposed. By leveraging mixed-norm regularisation, the locations of non-zero coefficients tend to cluster into groups. However, how to rationally constitute 'a group' is still a problem, since forcing local descriptors in a group to have uniform sparse pattern may be too rigid and bring about larger reconstruction error and information loss, especially when descriptors in a group are not very similar. In this paper, we propose multi-level K-means clustering and group sparse coding to address this issue, which is called MK-GSC. Moreover, we present a novel local descriptor based on SIFT to further promote the power of representation for images. Our method is tested on two public benchmarks, and achieves competitive or better results than the state-of-the-art methods.