Color Features Of Image Research Articles

Multi-Task Joint Sparse and Low-Rank Representation for the Scene Classification of High-Resolution Remote Sensing Image

Scene classification plays an important role in the intelligent processing of High-Resolution Satellite (HRS) remotely sensed images. In HRS image classification, multiple features, e.g., shape, color, and texture features, are employed to represent scenes from different perspectives. Accordingly, effective integration of multiple features always results in better performance compared to methods based on a single feature in the interpretation of HRS images. In this paper, we introduce a multi-task joint sparse and low-rank representation model to combine the strength of multiple features for HRS image interpretation. Specifically, a multi-task learning formulation is applied to simultaneously consider sparse and low-rank structures across multiple tasks. The proposed model is optimized as a non-smooth convex optimization problem using an accelerated proximal gradient method. Experiments on two public scene classification datasets demonstrate that the proposed method achieves remarkable performance and improves upon the state-of-art methods in respective applications.

Saliency Detection for Small Maritime Target Using Singular Value Decomposition of Amplitude Spectrum

ABSTRACTSpectrum analysis based saliency detection approaches have been studied and applied in computer vision. They are either generalized filtering or spectrum whitening of amplitude spectrum of image. However, as resultant salient regions may be inconsistent, a better result can be achieved by combining several scales. This paper proposes a novel saliency detection method for small maritime target using singular value decomposition (SVD) of amplitude spectrum. Color and intensity features of image are extracted and Fourier transform is employed on them. SVD is conducted on amplitude spectrums of each feature and their saliency maps are defined as the inverse Fourier transform of non-dominant components, whose amplitude spectrum is measured by difference between original amplitude spectrums and their main components. Taking into account the characteristics of maritime visual scene, the dominant parts of amplitude spectrum are selected according to threshold which is the ratio of singular values. Color saliency map and intensity saliency map are linearly combined to achieve the saliency map. The proposed method is conducted on the original image size, thus maintaining information of small maritime target. Experiments are made to show the validity and effectiveness. Comparison between proposed method and related approaches demonstrates that the new method is more suitable for saliency detection of small maritime target. Effect of threshold is also analyzed, and 50% is best for saliency detection of small maritime targets. The new saliency detection approach can be used for small maritime target detection and tracking.

A robust stereo matching algorithm based on improved epipolar distance transformation

Aim to the problem of initial epipolar distance transformation algorithm is sensitive to illumination and depth discontinuous, an improved stereo matching algorithm is proposed. In Delta-Gama log-chromaticity space, Gamma correction coefficient is computed firstly according to raw disparity map which is used to Modified the similarity function bandwidth. Secondly, by employing image structure and color features, the segment of epipolar is obtained adaptively. At last, final disparity map is acquired by belief propagation method. Experimental results demonstrate the effectiveness of the proposed transform in illumination change regions and discontinuities regions. The matching rate of proposed algorithm is improved about 60%, comparing with the initial epipolar distance transformation algorithm and state-of-art algorithm such as ANCC(Adaptive Normalized Cross Correlation).

Automated prediction of sensory scores for color and appearance in canned black beans (Phaseolus vulgaris L.) using machine vision

ABSTRACTEvaluation of canning quality of beans is commonly carried out by simple visual inspection that is time-consuming, resource intensive, and biased by the experience of the panelist. Moreover, there is not a standard scale to rate visual quality traits of canned beans. In this research, a machine vision system was implemented and tested for automatic inspection of color (COL) and appearance (APP) in canned black beans. Various color and textural image features (average, standard deviation, contrast, correlation, energy, and homogeneity from red, green, blue, lightness, red/green, yellow/blue, hue, saturation and value color scales) were extracted from beans and brine images, and evaluated to predict the quality rates for COL and APP of a group of bean panelists using multivariate statistics. Sixty-nine commercial canned black bean samples from different brands and markets were used for analysis. In spite of the “fair” agreement among the sensory panelists for COL and APP, as determined by multi-rater Kappa analysis, machine vision data based on partial least squares regression model showed high predictive performance for both COL and APP with correlation coefficients of 0.937 and 0.871, and standard errors of 0.26 and 0.38, respectively. When a classification was performed based on both COL and APP traits, a support vector machine model was able to sort the samples into two sensory quality categories of “acceptable” and “unacceptable” with an accuracy of 89.7%. Using simple color and texture image data, a machine vision system showed potential for the automatic evaluation of canned black beans by COL and/or appearance as a professional visual inspection.

A robust color image watermarking using local invariant significant bitplane histogram

Desynchronization attacks that cause displacement between embedding and detection are usually difficult for watermark to survive. It is a challenging work to design a robust image watermarking scheme against desynchronization attacks, especially for color images. In this paper, we propose a robust color image watermarking approach based on local invariant significant bitplane histogram. The novelty of the proposed approach includes: 1) A fast and effective color image feature points detector is constructed, in which probability density and color invariance model are used; 2) The fully affine invariant local feature regions are built based on probability density Hessian matrix; and 3) The invariant significant bitplane histograms are introduced to embed digital watermark. The extensive experimental works are carried out on a color image set collected from Internet, and the preliminary results show that the proposed watermarking approach can survive numerous kinds of distortions, including common image processing operations and desynchronization attacks.

깊이와 칼라 영상의 특징을 사용한 ROI 기반 객체 추출

깊이와 칼라 영상의 특징을 사용한 ROI 기반 객체 추출

Underexposed and overexposed colour image enhancement using information set theory

This paper presents a new approach for improving the contrast of digital colour images in extreme lighting environment. The concept of information sets that are derived from fuzzy sets is applied for this purpose. The S and V components of the HSV (hue, saturation, value) colour model are fuzzified using Gaussian and triangular membership functions but H is preserved. Objective measures like contrast information, quality and visual factor (Vf) are defined to represent the image characteristics. A new entropy function and Vf are used to form the objective function which is optimised using particle swarm optimisation to learn the image-specific parameters required for the enhancement. These new measures defined here serve to evaluate the performance of the proposed algorithm. In addition to preserving the colour and specific image features, the subjective and objective evaluations clearly depict the improvement in the quality of both the underexposed and overexposed images. On comparison, it is found that the proposed technique outperforms the existing techniques in terms of entropy, information measure and mean opinion score.

Automatic detection and severity measurement of eczema using image processing.

Chronic skin diseases like eczema may lead to severe health and financial consequences for patients if not detected and controlled early. Early measurement of disease severity, combined with a recommendation for skin protection and use of appropriate medication can prevent the disease from worsening. Current diagnosis can be costly and time-consuming. In this paper, an automatic eczema detection and severity measurement model are presented using modern image processing and computer algorithm. The system can successfully detect regions of eczema and classify the identified region as mild or severe based on image color and texture feature. Then the model automatically measures skin parameters used in the most common assessment tool called "Eczema Area and Severity Index (EASI)," by computing eczema affected area score, eczema intensity score, and body region score of eczema allowing both patients and physicians to accurately assess the affected skin.

Prediction of Pork Color Grade using Image Two-tone Color Ratio Features and Support Vector Machine

The objective of this study was to investigate the usefulness of pork loin color image features in predicting pork two-tone color grade according to objective L* value. Nine image color features (specifically, the means for two-tone ratios of R, G, B, L*, a*, b*, H, S and I) were extracted from 3 different color spaces (RGB (Red, Green and Blue), CIE LAB (L*: luminance; a*: green to red; b*: blue to yellow) and HIS (Hue, saturation and Intensity)). Color features were extracted from a laboratory-based high-quality camera imaging system. Objective color (CIE L*, a* and b*) was measured using a Minolta Colorimeter, calibrated using both white and black tiles. Boneless, 2.54-cm thick sirloin chops (enhanced, n = 541; non-enhanced, n = 232) were collected. K-means clustering technique was used for grouping pork into two color grades based on Minolta L* value. The image color features were used as predictors for multivariate classification of the samples using machine learning method (Support Vector Machine, SVM). For establishing the model, each data set was separated into training (70%) and testing (30%) sets. Ten-fold cross validation was used to set up the model and test for the best model parameters. The results showed that, for both enhanced and non-enhanced chops, the SVM machine method predicted 100% correct for both grades. Therefore, color image features can be used to correctly classify pork chops by SVM model according to the Minolta L* value.

Local quaternion PHT based robust color image watermarking algorithm

It is a challenging work to design a robust localized color image watermarking scheme against desynchronization attacks. There are two main drawbacks indwelled in current localized color image watermarking: firstly, the pure gray-based feature points detectors were utilized, in which the important color information is ignored. Secondly, the watermarking algorithms were designed mainly to mark the image luminance component only, in which the significant color channels correlation are neglected. In this paper, we propose a robust color image watermarking algorithm using local quaternion PHT (Polar Harmonic Transform), which is invariant to various noises, local geometric transformations, and color variations. Firstly, the stable color image feature points are extracted by using new color image feature point detector, in which the SIFER (Scale-Invariant Feature detector with Error Resilience) detector and color invariance model are incorporated. Then, the affine invariant local regions are built adaptively according to local image content variation. Finally, the digital watermark is embedded into the local regions by modulating the invariant quaternion PHT modulus coefficients. Experiments are carried out on a color image set collected from Internet, and the extensive experimental works have shown that the proposed color image watermarking is not only invisible and robust against common image processing operations such as median filtering, noise adding, and JPEG compression, but also has conquered those challenging desynchronization attacks.

Extraction of sparse features of color images in recognizing objects

In this paper, we propose a new object recognition framework that combines Gabor energy filters, a visual cortex model in which single units alternate with complex units, and color information. Each color image is first converted to the CIELAB color space. Rather using Gabor filters in the first layer of the cortex model, to each color component, a set of Gabor energy filters is applied. Thereafter, the superposition responses of the Gabor energy filter outputs over the color components are normalized by divisive normalization. In the fourth layer, sparse features are calculated using a localized pooling method that allows retention of some geometric information from the prototype patches’ positions. Finally, a set of sparse features are exploited by a linear SVM classifier for object recognition and classification. In the learning stage, a set of prototypes is selected randomly over spatial position, spatial size, and several scales simultaneously, and is extracted by the local maximum over scales and orientations, ignoring weaker training scales and orientations. The results of experiments performed on several datasets show that the use of color information in our framework improves object recognition significantly.

Classifier design based on the mechanism of capture-suppress of PCNN

针对医学图像识别的重要性, 本文提出了一种结合彩色图像特征场和基于脉冲耦合神经网络捕获-抑制机制的图像分类器设计方法. 首先, 构造符合人类视觉的彩色图像特征场; 然后, 利用脉冲耦合神经网络设计分类器实现对图像的识别. 最后, 应用于JAFFE人脸表情数据库和中医舌象的分类识别中, 并与ADABOOST方法和SVM方法进行比较. 实验结果表明, 本文方法可以获得更好的医学图像识别效果.

A method for discrimination of processed ginger based on image color feature and a support vector machine model

The image color feature was analysed using a support vector machine and principal component analysis to discriminate different processed products of ginger.

A Support Vector Machines based Classifications of the Diabetes Mellitus from Human Tongue Quantitative Features

This paper proposes a new non-invasive automatic method to detect and classify Diabetes Mellitus (DM) in the patients. The proposed automatic method incorporates tongue image processing and LS-SVM based classifier. Color, Texture and Geometry features of healthy and DM tongue images are extracted and analysed. The optimized values of the above feature are used to train the LS-SVM in order to detect and classify the DM automatically. The proposed methods are tested with a new database and the performance of the proposed method was tested with confusion matrix as it gives 85.38% accuracy.

Generalized Sign Recognition Based on the Gaussian Statistical Color Model for Intelligent Road Sign Inventory

The rapid progress in sensor and information technologies makes it possible to develop intelligent road sign inventory (IRSI) by automatically collecting and processing vehicle-borne multisensor data. This study developed a framework for IRSI with a vehicle-borne multisensor system. The IRSI framework consists of four modules: ( a) sensor vehicle design for multisensor data collection, ( b) sign recognition, ( c) sign data integration and attribute computation, and ( d) IRSI-based applications. Within the framework, sign recognition is the key to IRSI. The study proposed a novel Gaussian statistical color model (G-SCM) for image segmentation and color feature extraction. The G-SCM has good capability to model actual sign color distributions and has fast implementation for G-SCM-based color segmentation. With color features integrated into sign texture, shape, location, and other features, a novel generalized recognition algorithm is proposed for sign inventory. The proposed G-SCM model and the sign recognition method were tested with actual video log images. The experimental results show that the G-SCM can model actual sign colors and segment sign images accurately. The fast implementation can greatly enhance computation speed compared with existing color analysis methods. The developed sign recognition algorithm was tested to recognize all speed limit signs from video log images collected along a 140-km highway segment. The algorithm recognized 30 of 31 signs and achieved a 96.8% site recognition rate. Of the 136 images containing speed limit signs, the algorithm had only eight false positives. The results demonstrate that the proposed sign recognition method is promising for the development of IRSI.

Content Based Image Retrieval Using Gray Scale Weighted Average Method

High feature vector dimension quietly remained a curse element for Content Based Image Retrieval (CBIR) system which eventually degrades its efficiency while indexing similar images from database. This paper proposes CBIR system using Gray Scale Weighted Average technique for reducing the feature vector dimension. The proposed method is more suitable for color and texture image feature analysis as compared to color weighted average method as illustrated in literature review. To prove the effectiveness of retrieval system, two standard benchmark dataset namely, Wang and Amsterdam Library of Texture Images (A LOT) for color and texture have been selected to evaluate the system retrieval accuracies as well as efficiencies generated by each method. For the purpose of image similarity, Euclidean distance has been employed which matches query image feature vector with image database feature vectors. The experimental results generated by two methods showed that overall performance of the proposed method is relatively better in terms of average precision, average recall and its average retrieval time.

Completed local similarity pattern for color image recognition

In this paper, we propose a simple yet effective local color image descriptor, completed local similarity pattern (CLSP), for face recognition. We represent the color image as the co-occurrence of its image pixel color quantization information and the local color image textural information. Specifically, CLSP consists of two complementary components: color label and local similarity pattern. Expressed by soft color label for every image pixel, we adopt the k-means clustering method and the soft-assignment coding method to summarize and encode the image pixel color information globally ignoring local neighboring pixels. While based on the similarity of color information between central pixel and its neighbors, local similarity pattern (LSP) is used to encode the local spatial textural feature of the color image ignoring their color value. Therefore, LSP has the merits of robustness and compactness of coding based feature extraction method. The joint distribution (2-D histogram), which unifies the color label and LSP, is used for a given image representation. Moreover, seven different color spaces are selected and fused to compensate each other for color image feature extraction. Experimental results on GeorgiaTech, AR, NUST-RWFR and LFW face database show that CLSP outperforms state-of-the-art color image feature extraction methods.

People recognition in multi-cameras using the visual color processing mechanism

In this paper a people recognition algorithm is proposed, in which the color processing mechanism is inspired by the biological visual system. The algorithm is constructed in two parts. In the first part a spiking neural network is proposed to extract the color features of the people images which are captured from videos, and a set of new features is generated by fusing the color features and color moments. In the second part, after a feature reduction, a Support Vector Machine is trained and then used to recognize a specific people. The algorithm has been successfully applied to recognize people in CASIA Database with a high recognition rate. In order to evaluate performance and analyze characteristics of people recognition algorithms in multi-camera scenes, Multi-Camera Video (MCV) dataset is made in this paper. It is used to evaluate and analyze the proposed method and a set of characteristics of the proposed algorithms are obtained. Experimental results demonstrate that the algorithm is comparable to state-of-the-art approaches in terms of accuracy and the direction for further improvement of the proposed algorithm is provided.

Predicting pork two tone color grade using image color features and support vector machine

Predicting pork two tone color grade using image color features and support vector machine

Prediction of pork color attributes using computer vision system

Color image processing and regression methods were utilized to evaluate color score of pork center cut loin samples. One hundred loin samples of subjective color scores 1 to 5 (NPB, 2011; n=20 for each color score) were selected to determine correlation values between Minolta colorimeter measurements and image processing features. Eighteen image color features were extracted from three different RGB (red, green, blue) model, HSI (hue, saturation, intensity) and L*a*b* color spaces. When comparing Minolta colorimeter values with those obtained from image processing, correlations were significant (P<0.0001) for L* (0.91), a* (0.80), and b* (0.66). Two comparable regression models (linear and stepwise) were used to evaluate prediction results of pork color attributes. The proposed linear regression model had a coefficient of determination (R2) of 0.83 compared to the stepwise regression results (R2=0.70). These results indicate that computer vision methods have potential to be used as a tool in predicting pork color attributes.