Search

To solve the problem that the traditional hyperspectral image classification method cannot effectively distinguish the boundary of objects with a single scale feature, which leads to low classification accuracy, this paper introduces the idea of guided filtering into hyperspectral image classification, and then proposes a multi-scale guided feature extraction and classification (MGFEC) algorithm for hyperspectral images. Firstly, the principal component analysis theory is used to reduce the dimension of hyperspectral image data. Then, guided filtering algorithm is used to achieve multi-scale spatial structure extraction of hyperspectral image by setting different sizes of filtering windows, so as to retain more edge details. Finally, the extracted multi-scale features are input into the support vector machine classifier for classification. Several practical hyperspectral image datasets were used to verify the experiment, and compared with other spectral feature extraction algorithms. The experimental results show that the multi-scale features extracted by the MGFEC algorithm proposed in this paper are more accurate than those extracted by only using spectral information, which leads to the improvement of the final classification accuracy. This fully shows that the proposed method is not only effective, but also suitable for processing different hyperspectral image data. [ABSTRACT FROM AUTHOR]

Analyzing geological drilling hole images acquired by Axial View Panoramic Borehole Televiewer (APBT) is a key step to explore the geological structure in a geological exploration. Conventionally, the borehole images are examined by technicians, which is inefficient and subjective. In this paper, three dominant types of borehole-wall images on coal-rock mass structure, namely, border images, fracture images and intact rock mass images are mainly studied. The traditional image classification methods based on unified feature extraction algorithm and single classifier is not effect for the borehole images. Therefore, this paper proposes a novel two-stage classification approach to improve the classification performance of borehole images. In the first-stage classification, the border images are identified from three kinds of images based on texture features and gray-scale histograms features. For the remaining two types of images, in the second-stage classification, Gabor filter is first applied to segment the region of interest (ROI) (such as microfracture, absciss layer and horizontal cracks, etc.) and the central interference region. Then, using the same feature vector after eliminating the central interference region, fracture images are separated from intact rock mass images. We test our two-stage classification system with real borehole images. The results of experimental show that the two-stage classification method can effectively classify three major borehole-wall images with the correction rate of 95.55% in the first stage and 95% in the second stage. [ABSTRACT FROM AUTHOR]

This paper attempts to design an automated, efficient and intelligent tomato grading method that facilitates the graded selling of the fruit. Based on machine vision, the color images of tomatoes with different morphologies were studied, and the color, shape and size were selected as the key features. On this basis, an automated grading classifier was created based on the surface features of tomatoes, and a grading platform was set up to verify the effect of the classifier. Specifically, the Hue value distributions of tomatoes with different maturities were investigated, and the Hue value ranges were determined for mature, semi-mature and immature tomatoes, producing the color classifier. Next, the first-order Fourier descriptor (1D- FD) was adopted to describe the radius sequence of tomato contour, and an equation was established to compute the irregularity of tomato contour, creating the shape classifier. After that, a linear regression equation was constructed to reflect the relationship between the transverse diameters of actual tomatoes and tomato images, and a classifier between large, medium and small tomatoes was produced based on the transverse diameter. Finally, a comprehensive tomato classifier was built based on the color, shape and size diameters. The experimental results show that the mean grading accuracy of the proposed method was 90.7%. This means our method can achieve automated real-time grading of tomatoes. [ABSTRACT FROM AUTHOR]

Recognition and classification for wire bonding joint are important to quality assurance in semiconductor device manufacturing. In this paper, a precision recognition and classification system for bonding joint of ultrasonic heavy aluminum wire based on image feature and support vector machine (SVM) is presented. This system consists of feature extraction from images and classification model. In feature extraction, image processing algorithms including Canny edge extraction, histogram equalization, and image morphology closed operation are utilized to extract and locate a joint contour in a complicated background image. In the classification model, the principal component analysis (PCA) is employed to visualize, reconstruct, and reduce the images data dimension for less computation time. The SVM-based model is chosen as the classifier to identify and recognize joint types. The Gauss-radial basis function (RBF) kernel function is adopted in SVM, and its optimal parameters are determined by cross-validation. In the experiment, 588 bonding images are used to implement in this recognition and classification system. The results prove that the classification accuracy for wire bonding joint based on image feature, PCA, and SVM can achieve to 97.3%, and the computation time can be reduced significantly. [ABSTRACT FROM AUTHOR]

Deciding a rational number of clusters for the problems to be solved and determining the parameters associated with the clustering algorithms are two critical issues in the configuration of data clustering. Usually, a manual trial-and-error manner is used to induce a feasible configuration. This paper presents an automated clustering method, which determines the clustering configuration automatically. The proposed method is based on the techniques of support-vector machines with a local-search strategy. It starts with running one-class support-vector machines (OCSVM) to partition input data into a random number of clusters. When a result is obtained, the “local-search” mechanism launches several rounds of OCSVM each of which works with a new clustering configuration. Each new configuration is from the current configuration with incrementally modifications. The clustering results obtained from the local searches are post-evaluated by specific clustering validity index and the best one is retained. The clustering configuration of the best result is used by OCSVM for the clustering afterwards. Such a clustering process iterates until no better result can be obtained. This paper describes the clustering algorithm and compares three clustering validity indices, i.e. distance-based index, Davies–Bouldin index, and Xie–Beni index, on their effectiveness. The performance of the proposed method is demonstrated on the segmentation of several aerial images. Experimental results show that the proposed approach is feasible and effective for image segmentation. [ABSTRACT FROM AUTHOR]

During the detecting and locating process of green tomatoes based on obtained visible images, problems such as shadow caused by uneven illumination source, occlusion of stems and leaves, and occlusion between fruits, need to be solved. In this study, a machine vision algorithm was put forward, which aimed to determine fruit location and size of green tomatoes. Normalized cross-correlation function (NCC) is a feature detection method using template matching. The algorithm firstly detected the potential location of green tomatoes through fast normalized cross-correlation function (FNCC). Then a gray histogram based classifier was used to classify if the location corresponded to green fruit. The histogram based classifier was built based on fruit areas and non-fruit areas of a certain size (30×30 pixels) extracted from the obtained image. Seven texture features, including mean, standard deviation, smoothness, third moment, uniformity, entropy, and gray level range were calculated for those fruit areas and non-fruit areas. Three classifiers including k-nearest neighbor (KNN), SVM (support vector machine), and Naive Bayes, were used to classify fruit and non-fruit areas using those 7 texture features as input vectors. SVM was chosen based on its performance. The non-fruit location was filtered out, and the number of fruit locations was estimated. Meanwhile, the image was segmented based on color analysis. Red and Blue component from RGB (red, green, blue) image, and Hue component from HSV (hue, saturation, value) image transformed from RGB images, were used as the basis for color analysis. Using the fruit potential location number estimated using FNCC as an input parameter of circular Hough transform (CHT), CHT was then applied to the edge image of the segmented result. The center coordinates and radius value of each circle were calculated. Finally, the detection results were merged based on the analysis of the distances of 2 centers of fruit circles detected using CHT. If the distance between 2 circles is smaller than the minimum fruit radius, the circle with a larger radius will be kept, while the other circle will be flagged as repeatedly detected one. Thus, the recognition and positioning of the green tomato were realized. When green fruits and red fruits appear on the same image, a red fruit detection algorithm based on the local maximum value method and random circle round transform detection, which was proposed by the author in another paper, would be carried out on the obtained image. Then the red fruit detection result was combined with the green fruit detection result. The proposed method combined the texture, color, and shape information of the image, and eliminated the disturbance of the color similarity between the green tomatoes and green leaves and stems. A total of 70 images were used in this study, in which 35 images were used as training images, and the other 35 images were used as validation images. The correct detection ratio for 72 fruits in the training dataset was 89.2%, and the correct detection ratio for 105 fruits in the validation dataset was 86.7%. The proposed method has provided a reference for the development of tomato harvesting robots for both red and green mature tomato fruits. [ABSTRACT FROM AUTHOR]

Full rank principal component analysis (FR-PCA) is a special form of principal component analysis (PCA) which retains all nonzero components of PCA. Generally speaking, it is hard to estimate how the accuracy of a classifier will change after data are compressed by PCA. However, this paper reveals an interesting fact that the transformation by FR-PCA does not change the accuracy of many well-known classification algorithms. It predicates that people can safely use FR-PCA as a preprocessing tool to compress high-dimensional data without deteriorating the accuracies of these classifiers. The main contribution of the paper is that it theoretically proves that the transformation by FR-PCA does not change accuracies of the k nearest neighbor, the minimum distance, support vector machine, large margin linear projection, and maximum scatter difference classifiers. In addition, through extensive experimental studies conducted on several benchmark face image databases, this paper demonstrates that FR-PCA can greatly promote the efficiencies of above-mentioned five classification algorithms in appearance-based face recognition. [ABSTRACT FROM AUTHOR]

In this paper, a novel Multiview CLOUD (mCLOUD) visual feature extraction mechanism is proposed for the task of categorizing clouds based on ground-based images. To completely characterize the different types of clouds, mCLOUD first extracts the raw visual descriptors from the views of texture, structure, and color simultaneously, in a densely sampled way-specifically, the scale invariant feature transform (SIFT), the census transform histogram (CENTRIST), and the statistical color features are extracted, respectively. To obtain a more descriptive cloud representation, the feature encoding of the raw descriptors is realized by using the Fisher vector. This is followed by the feature aggregation procedure. A linear support vector machine (SVM) is employed as the classifier to yield the final cloud image categorization result. The experiments on a challenging cloud dataset termed the six-class Huazhong University of Science and Technology (HUST) cloud demonstrate that mCLOUD consistently outperforms the state-of-the-art cloud classification approaches by large margins (at least 6.9%) under all the different experimental settings. It has also been verified that, compared to the single view, the multiview cloud representation generally enhances the performance. [ABSTRACT FROM AUTHOR]

The Chagas disease is a potentially life-threatening illness caused by the protozoan parasite, Trypanosoma cruzi. Visual detection of such parasite through microscopic inspection is a tedious and time-consuming task. In this paper, we provide an AdaBoost learning solution to the task of Chagas parasite detection in blood images. We give details of the algorithm and our experimental setup. With this method, we get 100% and 93.25% of sensitivity and specificity, respectively. A ROC comparison with the method most commonly used for the detection of malaria parasites based on support vector machines (SVM) is also provided. Our experimental work shows mainly two things: (1) Chagas parasites can be detected automatically using machine learning methods with high accuracy and (2) AdaBoost + SVM provides better overall detection performance than AdaBoost or SVMs alone. Such results are the best ones known so far for the problem of automatic detection of Chagas parasites through the use of machine learning, computer vision, and image processing methods. [ABSTRACT FROM AUTHOR]

In this paper, we propose a matrix version extension for linear regularization projection twin support vector machine presented by Shao et al. (Knowl Based Syst 37:203-210, ), named as linear projection twin support matrix machine [linear projection twin support matrix machine (PTSMM)], for 2d image data classification. In order to discuss the nonlinear version of PTSMM, a new matrix kernel function is introduced and based on which, we provide a nonlinear PTSMM algorithm with a detailed theoretical derivation. To examine the effectiveness of the presented linear and nonlinear PTSMM, we perform comparative experiments with three linear classifiers support tensor machines, twin support tensor machine and proximal support tensor machine on ORL, YALE and AR databases. Experimental results show that our methods are effective and efficient. [ABSTRACT FROM AUTHOR]

Synthetic fingerprints are a potential threat to automatic fingerprint identification systems (AFISs). In this paper, we propose an algorithm to discriminate synthetic fingerprints from real ones. First, four typical characteristic factors—the ridge distance features, global gray features, frequency feature and Harris Corner feature—are extracted. Then, a support vector machine (SVM) is used to distinguish synthetic fingerprints from real fingerprints. The experiments demonstrate that this method can achieve a recognition accuracy rate of over 98% for two discrete synthetic fingerprint databases as well as a mixed database. Furthermore, a performance factor that can evaluate the SVM's accuracy and efficiency is presented, and a quantitative optimization strategy is established for the first time. After the optimization of our synthetic fingerprint discrimination task, the polynomial kernel with a training sample proportion of 5% is the optimized value when the minimum accuracy requirement is 95%. The radial basis function (RBF) kernel with a training sample proportion of 15% is a more suitable choice when the minimum accuracy requirement is 98%. [ABSTRACT FROM AUTHOR]

Existing research works in scene image classification have focused on different aspects such as content features (e.g., visual information), context features (e.g., annotations, semantic information, etc.) and both. However, such works suffer from various issues such as higher feature size and lower classification performance. In this paper, we propose a new feature extraction approach for scene image representation using two kinds of rich information: content features and context features. Specifically, the new content features are generated by multi-scale foreground and background information. Similarly, the new context features are generated by the novel compact supervised codebook. Our compact supervised codebook minimizes irrelevant and redundant information, which, in result, achieves the lower-sized contextual feature vector. Finally, we combine both content and context features to represent the scene image. Our experiments on three widely used benchmark scene datasets using Support Vector Machine (SVM) classifier reveal that our proposed context and content features produce better results than existing context and content features, respectively. The fusion of the proposed two types of features significantly outperform numerous state-of-the-art features. • Propose a novel contextual feature extraction method for scene images. • Propose a novel unique tokens extraction algorithm. • Aggregate multi-scale content and contextual features to represent the scene images. • Evaluate our proposed method on three commonly-used benchmark datasets. [ABSTRACT FROM AUTHOR]

Currently, remote sensing technologies were widely employed in the dynamic monitoring of the land. This paper presented an algorithm named fuzzy nonlinear proximal support vector machine (FNPSVM) by basing on ETM+ remote sensing image. This algorithm is applied to extract various types of lands of the city Da’an in northern China. Two multi-category strategies, namely “one-against-one” and “one-against-rest” for this algorithm were described in detail and then compared. A fuzzy membership function was presented to reduce the effects of noises or outliers on the data samples. The approaches of feature extraction, feature selection, and several key parameter settings were also given. Numerous experiments were carried out to evaluate its performances including various accuracies (overall accuracies and kappa coefficient), stability, training speed, and classification speed. The FNPSVM classifier was compared to the other three classifiers including the maximum likelihood classifier (MLC), back propagation neural network (BPN), and the proximal support vector machine (PSVM) under different training conditions. The impacts of the selection of training samples, testing samples and features on the four classifiers were also evaluated in these experiments. [ABSTRACT FROM AUTHOR]

This paper presents a high-speed image processing algorithm applied to multifaceted asymmetric radiation from the edge detection on Joint European Torus. The algorithm has been developed in C/C++ language using the OpenCV library for image processing and the LibSVM library for pattern classification and has been tested on a dedicated Intel 64-bit Linux computing platform. The code implemented achieves at the same time high accuracy and good run-time performance. The final version presents a correct detection rate of 93.3\% and an average image processing rate of 650 frames per second. A complete analysis of each image processing module is presented to illustrate the overall characteristics and performance of the algorithm. [ABSTRACT FROM PUBLISHER]

Machine learning techniques for computer vision applications like object recognition, scene classification, etc., require a large number of training samples for satisfactory performance. Especially when classification is to be performed over many categories, providing enough training samples for each category is infeasible. This paper describes new ideas in multiclass active learning to deal with the training bottleneck, making it easier to train large multiclass image classification systems. First, we propose a new interaction modality for training which requires only yes-no type binary feedback instead of a precise category label. The modality is especially powerful in the presence of hundreds of categories. For the proposed modality, we develop a Value-of-Information (VOI) algorithm that chooses informative queries while also considering user annotation cost. Second, we propose an active selection measure that works with many categories and is extremely fast to compute. This measure is employed to perform a fast seed search before computing VOI, resulting in an algorithm that scales linearly with dataset size. Third, we use locality sensitive hashing to provide a very fast approximation to active learning, which gives sublinear time scaling, allowing application to very large datasets. The approximation provides up to two orders of magnitude speedups with little loss in accuracy. Thorough empirical evaluation of classification accuracy, noise sensitivity, imbalanced data, and computational performance on a diverse set of image datasets demonstrates the strengths of the proposed algorithms. [ABSTRACT FROM AUTHOR]

Abstract: Remote diagnose of fish diseases for farmers is unrealized in China, but use of mobile phones and remote analysis based on image processing can be feasible due to the widespread use of mobile phones with camera features in rural areas. This paper presents a novel method of classifying species of fish based on color and texture features and using a multi-class support vector machine (MSVM). Fish images were acquired and sent by smartphone, and the method utilized was comprised of the following stages. Color and texture subimages of fish skin were obtained from original images. Color features, statistical texture features and wavelet-based texture features of the color and texture subimages were extracted, and six groups of feature vectors were composed. LIBSVM software was tested using leave-one-out cross validation to find the best group for classification in feature selection procedure. Two multi-class support vector machines based on a one-against-one algorithm were constructed for classification. The feature selection results showed that the Bior4.4 wavelet filter in HSV color space achieved greater accuracy than the other feature groups. The classification results indicate that only the DAGMSVM meets the requirement of time efficiency for the system. The results of this study suggest that the best classification model for fish species recognition is composed of a wavelet domain feature extractor with Bior4.4 wavelet filter in HSV color space and a one-against-one algorithm based DAGMSVM classifier. [Copyright &y& Elsevier]

Abstract: This paper presents new methods for an automated analysis of the double InterTropical Convergence Zone (dITCZ) phenomena on a daily time scale over the east Pacific. Long-term Geostationary Operational Environmental Satellite (GOES) visible and infrared data are used to spatially identify and segment the convection zones over the east Pacific basin on both sides of the equator and to track the temporal variability of the ITCZ, specifically to identify cases of dITCZs, northern or southern ITCZ, or non-presence events. For the segmentation approach, image processing techniques are developed to extract information about the spatial features of the ITCZ in both hemispheres for each satellite image. These features serve as input to a temporal classification algorithm that is based on a combination of hidden semi Markov model (HsMM) and support vector machine (SVM) methods. The performance of the proposed method is competitive with human experts and the methodology can thus be used to conduct an in-depth analysis of the dITCZ. Such an analysis could provide precise information for refining existing weather and climate models over the sparsely observed east Pacific where the dITCZ is greatly over-represented in most models. [Copyright &y& Elsevier]

The task of face recognition has been actively researched in recent years because of its many applications in various domains. This paper presents a robust face recognition system using curvelet-based two-dimensional principle component analysis (2D PCA) to address the problem of human face recognition from still images. 2D PCA has advantages over PCA in evaluating the covariance matrix accurately and time complexity. Inspired by the attractive attributes of curvelets in catching the edge singularities with very few coefficients in a non-adaptive manner, we introduce the scheme of decomposing images into curvelet subbands and applying 2D PCA to create a representative feature set. Experiments were designed with different implementations of each module using standard testing database. We experimented with changing the illumination normalization procedure; comparing the baseline PCA-based method with the proposed scheme; studying effects on algorithm performance of k-nearest neighbor (kNN) classifier and Support Vector Machine (SVM) classifier in the classification process; also we experimented with different databases such as FERET, etc. High accuracy rate were achieved by the proposed scheme through a comparative study. [ABSTRACT FROM AUTHOR]

The goal of feature selection is to identify the most informative features for compact representation, whereas the goal of active learning is to select the most informative instances for prediction. Previous studies separately address these two problems, despite of the fact that selecting features and instances are dual operations over a data matrix. In this paper, we consider the novel problem of simultaneously selecting the most informative features and instances and develop a solution from the perspective of optimum experimental design. That is, by using the selected features as the new representation and the selected instances as training data, the variance of the parameter estimate of a learning function can be minimized. Specifically, we propose a novel approach, which is called Unified criterion for Feature and Instance selection (UFI), to simultaneously identify the most informative features and instances that minimize the trace of the parameter covariance matrix. A greedy algorithm is introduced to efficiently solve the optimization problem. Experimental results on two benchmark data sets demonstrate the effectiveness of our proposed method. [ABSTRACT FROM PUBLISHER]

Abstract: In this paper we propose a machine learning approach to classify melanocytic lesions as malignant or benign, using dermoscopic images. The lesion features used in the classification framework are inspired on border, texture, color and structures used in popular dermoscopy algorithms performed by clinicians by visual inspection. The main weakness of dermoscopy algorithms is the selection of a set of weights and thresholds, that appear not to be robust or independent of population. The use of machine learning techniques allows to overcome this issue. The proposed method is designed and tested on an image database composed of 655 images of melanocytic lesions: 544 benign lesions and 111 malignant melanoma. After an image pre-processing stage that includes hair removal filtering, each image is automatically segmented using well known image segmentation algorithms. Then, each lesion is characterized by a feature vector that contains shape, color and texture information, as well as local and global parameters. The detection of particular dermoscopic patterns associated with melanoma is also addressed, and its inclusion in the classification framework is discussed. The learning and classification stage is performed using AdaBoost with C4.5 decision trees. For the automatically segmented database, classification delivered a specificity of 77% for a sensitivity of 90%. The same classification procedure applied to images manually segmented by an experienced dermatologist yielded a specificity of 85% for a sensitivity of 90%. [Copyright &y& Elsevier]

Reliably detecting objects in broad-area overhead or satellite images has become an increasingly pressing need, as the capabilities for image acquisition are growing rapidly. The problem is particularly difficult in the presence of large intraclass variability, e.g., finding “boats” or “buildings,” where model-based approaches tend to fail because no good model or template can be defined for the highly variable targets. This paper explores an automatic approach to detect and classify targets in high-resolution broad-area satellite images, which relies on detecting statistical signatures of targets, in terms of a set of biologically-inspired low-level visual features. Broad-area images are cut into small image chips, analyzed in two complementary ways: “attention/saliency” analysis exploits local features and their interactions across space, while “gist” analysis focuses on global nonspatial features and their statistics. Both feature sets are used to classify each chip as containing target(s) or not, using a support vector machine. Four experiments were performed to find “boats” (Experiments 1 and 2), “buildings” (Experiment 3) and “airplanes” (Experiment 4). In experiment 1, 14 416 image chips were randomly divided into training (300 boat, 300 nonboat) and test sets (13 816), and classification was performed on the test set (ROC area: 0.977 \pm 0.003). In experiment 2, classification was performed on another test set of 11 385 chips from another broad-area image, keeping the same training set as in experiment 1 (ROC area: 0.952 \pm 0.006). In experiment 3, 600 training chips (300 for each type) were randomly selected from 108 885 chips, and classification was conducted (ROC area: 0.922 \pm 0.005). In experiment 4, 20 training chips (10 for each type) were randomly selected to classify the remaining 2581 chips (ROC area: 0.976 \pm 0.003). The proposed algorithm outperformed the state-of-the-art SIFT, HMAX, and hidden-scale salient structure methods, and previous gist-only features in all four experiments. This study shows that the proposed target search method can reliably and effectively detect highly variable target objects in large image datasets. [ABSTRACT FROM AUTHOR]

Abstract: This paper proposes image processing algorithms to recognize five types of white blood cells in peripheral blood automatically. First, a method based on Gram–Schmidt orthogonalization is proposed along with a snake algorithm to segment nucleus and cytoplasm of the cells. Then, a variety of features are extracted from the segmented regions. Next, most discriminative features are selected using a Sequential Forward Selection (SFS) algorithm and performances of two classifiers, Artificial Neural Network (ANN) and Support Vector Machine (SVM), are compared. The results demonstrate that the proposed methods are accurate and sufficiently fast to be used in hematological laboratories. [Copyright &y& Elsevier]

Abstract: This paper describes the designing and testing process of a vision system for strongly reflected metal’s surface defects detection. In the authors’ view, an automatic inspection system has the following stages: image acquisition, image pre-processing, feature extraction and classification. Thus, the system including four subsystems is designed, and image processing method and pattern recognition algorithm that perform specific functions are outlined. First, the study uses wavelet smoothing method to eliminate noise from the images. Then, the images are segmented by Otsu threshold. At last, five characteristics based on spectral measure of the binary images are collected and input into a support vector machine (SVM). Furthermore, kernel function selection and parameters settings which are used for SVM method are evaluated and discussed. Also, a very difficult detection case for the surface defects of strongly reflected metal is depicted in details. The classification results demonstrate that the proposed method can identify seven classes of metal surface defects effectively, and the results are summarized and interpreted. [ABSTRACT FROM AUTHOR]

Abstract: While Histograms of Oriented Gradients (HOG) plus Support Vector Machine (SVM) (HOG+SVM) is the most successful human detection algorithm, it is time-consuming. This paper proposes two ways to deal with this problem. One way is to reuse the features in blocks to construct the HOG features for intersecting detection windows. Another way is to utilize sub-cell based interpolation to efficiently compute the HOG features for each block. The combination of the two ways results in significant increase in detecting humans—more than five times better. To evaluate the proposed method, we have established a top-view human database. Experimental results on the top-view database and the well-known INRIA data set have demonstrated the effectiveness and efficiency of the proposed method. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, a new hierarchical classification method based on the use of various types of AdaBoost classification algorithms is proposed for automatic classification of marble slab images according to their quality. At first, features are extracted using the sum and difference histograms method and, at the second stage, different versions of the AdaBoost algorithms are used as classifiers together with those extracted features in a proposed hierarchical fashion. Performance of the proposed method is compared against performances of different types of neural network classifiers and a support vector machine (SVM) classifier. Computational results show that the proposed hierarchical structure employing AdaBoost algorithms performs superior to neural networks and the SVM classifier for classifying marble slab images in our large and diversified data set. [ABSTRACT FROM AUTHOR]

Abstract: The problem described in this paper consists in re-identifying moving people in different sites which are completely covered with non-overlapping cameras. Our proposed framework relies on the spectral classification of the appearance-based signatures extracted from the detected person in each sequence. We first propose a new feature called “color-position” histogram combined with several illumination invariant methods in order to characterize the silhouettes in static images. Then, we develop an algorithm based on spectral analysis and support vector machines (SVM) for the re-identification of people. The performance of our system is evaluated on real datasets collected on INRETS premises. The experimental results show that our approach provides promising results for security applications. [Copyright &y& Elsevier]

Biologically inspired feature (BIF) and its variations have been demonstrated to be effective and efficient for scene classification. It is unreasonable to measure the dissimilarity between two BIFs based on their Euclidean distance. This is because BIFs are extrinsically very high dimensional and intrinsically low dimensional, i.e., BIFs are sampled from a low-dimensional manifold and embedded in a high-dimensional space. Therefore, it essential to find the intrinsic structure of a set of BIFs, obtain suitable mapping to implement the dimensionality reduction, and measure the dissimilarity between two BIFs in the low-dimensional space based on their Euclidean distance. In this paper, we study the manifold constructed by a set of BIFs utilized for scene classification, form a new dimensionality reduction algorithm by preserving both the geometry of intra BIFs and the discriminative information inter BIFs termed Discriminative and Geometry Preserving Projections (DGPP), and construct a new framework for scene classification. In this framework, we represent an image based on new BIF, which combines the intensity channel, the color channel, and the C1 unit of a color image; then we project the high-dimensional BIF to a low-dimensional space based on DGPP; and, finally, we conduct the classification based on the multiclass support vector machine (SVM). Thorough empirical studies based on the USC scene dataset demonstrate that the proposed framework improves the classification rates around 100% relatively and the training speed 60 times for different sites in comparing with previous gist proposed by Siagian and Itti in 2007. [ABSTRACT FROM AUTHOR]

Many types of nonlinear classifiers have been proposed to automatically generate land-cover maps from satellite images. Some are based on the estimation of posterior class probabilities, whereas others estimate the decision boundary directly. In this paper, we propose a modular design able to focus the learning process on the decision boundary by using posterior probability estimates. To do so, we use a self-configuring architecture that incorporates specialized modules to deal with conflicting classes, and we apply a learning algorithm that focuses learning on the posterior probability regions that are critical for the performance of the decision problem stated by the user-defined misclassification costs. Moreover, we show that by filtering the posterior probability map, the impulsive noise, which is a common effect in automatic land-cover classification, can be significantly reduced. Experimental results show the effectiveness of the proposed solutions on real multi- and hyperspectral images, versus other typical approaches, that are not based on probability estimates, such as Support Vector Machines. [ABSTRACT FROM AUTHOR]

The quality of pizza base should be determined before further processing. However such classification is highly sensitive to human error. Image processing techniques can be used to extract shape features from digital images of pizza bases. The support vector machine (SVM) is the state-of-the-art classification technique and is capable of learning in high-dimensional feature space with less training data. In this paper, an algorithm for the shape extraction of pizza base and a new classification algorithm based on Fourier transform and SVM for shape grading of pizza bases are presented. The experiments showed that 86.7% classification accuracy was achieved with a linear SVM classifier, 95.0% with a polynomial SVM classifier, and 98.3% with a Gaussian radial basis function SVM classifier. The computer vision system developed has a great potential to assist in the automation of pizza base classification. [Copyright &y& Elsevier]

Recently, networks consider spectral-spatial information in multiscale inputs less, even though there are some networks that consider this factor, however these networks cannot guarantee to get optimal features, which are extracted from each scale input. Furthermore, these networks do not consider the complementary and related information among different scale features. To address these issues, a multiscale deep middle-level feature fusion network (MMFN) is proposed in this paper for hyperspectral classification. In MMFN, the network fully fuses the strong complementary and related information among different scale features to extract more discriminative features. The training of network contains two stages: the first stage obtains the optimal models corresponding to different scale inputs and extracts the middle-level features under the corresponding scale model. It can guarantee the multiscale middle-level features are optimal. The second stage fuses the optimal multiscale middle-level features in the convolutional layer, and the subsequent residual blocks can learn the complementary and related information among different scale middle-level features. Moreover, the idea of identity mapping in residual learning can help the network obtain a higher accuracy when the network is deeper. The effectiveness of our method is proved on four HSI data sets and the experimental results show that our method outperforms the other state-of-the-art methods especially with small training samples. [ABSTRACT FROM AUTHOR]