The problem of image feature extraction for classification is difficult because of the high dimensionality inherent in image data. By extracting only relevant image features we reduce the dimensionality of the problem...
详细信息
ISBN:
(纸本)0819425915
The problem of image feature extraction for classification is difficult because of the high dimensionality inherent in image data. By extracting only relevant image features we reduce the dimensionality of the problem and improve classification accuracy. We further enhance classification performance by finding an optimal representation of the extracted image features which maximizes separability distance among classes. The principal tools used are Fourier series, wavelet packets, local discriminant basis analysis, and neural networks.
In progress towards automatic cancer pre-screening, it is essential to achieve successful image segmentation prior to cancer pre-screening. In this paper, a medical image segmenter system, which obtains local informat...
详细信息
ISBN:
(纸本)085296692X
In progress towards automatic cancer pre-screening, it is essential to achieve successful image segmentation prior to cancer pre-screening. In this paper, a medical image segmenter system, which obtains local information from Artificial Life Region Labelling and global information using mathematical morphology is described. Also, we describe progress towards an automatic segmentation system that employs objective segmentation measures.
In this work we present the application of different indexes based on the modal matching technique to visual search in an image database. The problem of efficiently retrieving the images similar to a user-defined sket...
详细信息
ISBN:
(纸本)085296692X
In this work we present the application of different indexes based on the modal matching technique to visual search in an image database. The problem of efficiently retrieving the images similar to a user-defined sketch is addressed. Similarity evaluation for scarcely sampled shapes is outlined, as well as the problems related to modal matching between differently sampled objects. To these aims, four different definitions of suitable similarity indexes are introduced and discussed.
The effectiveness of using region-based feature data as keys to an image database query search is presented. The classification engine required for this form of query has been implemented using a radial-basis function...
详细信息
ISBN:
(纸本)085296692X
The effectiveness of using region-based feature data as keys to an image database query search is presented. The classification engine required for this form of query has been implemented using a radial-basis function (RBF) network. It provides significant results using low volume training data provided in the form of feedback from the user on the progress of the query.
Genetic algorithms provide a valuable method for the optimization and the synthesis of (near) optimal filters for image noise removal. A method that does not use the original image is proposed. The filter is synthesiz...
详细信息
ISBN:
(纸本)085296692X
Genetic algorithms provide a valuable method for the optimization and the synthesis of (near) optimal filters for image noise removal. A method that does not use the original image is proposed. The filter is synthesized on a small (8 by 8 pixels) artificial image which contains the representative building blocks of a natural image (flat regions and edges). This test image is degraded by the same (or a similar) noise as the noise affecting the natural image. The genetic algorithm is used for optimizing the 3 by 3 point's window L-filter that obtains the best results in terms of mean square errors and MAE.
In this paper, a novel approach to feature extraction for rotationally invariant object classification is proposed based directly on a discrete wavelet transformation. This form of feature extraction is equivalent to ...
详细信息
ISBN:
(纸本)0819424838
In this paper, a novel approach to feature extraction for rotationally invariant object classification is proposed based directly on a discrete wavelet transformation. This form of feature extraction is equivalent to retaining informative features while eliminating redundant features from images, which is a critical property when analyzing large, high dimensional images. Usually, researches have resorted to a data pre-processing method to reduce the size of the feature space prior to classification. The proposed method employs statistical features extracted directly from the wavelet coefficients generated from a three-level subband decomposition system using a set of orthogonal and regular quadrature mirror filters [1], [13]. This algorithm has two desirable properties: 1) It reduces the number of dimensions of the feature space necessary to achieve the same classification accuracy as in the original space for a given pattern recognition problem;2) regardless of the target orientation, the algorithm can perform classification with low error rates. Furthermore, the filters used have performed well in the image compression regime [6],[10],[11], but they have not been applied to applications in target classification which will be demonstrated in this paper. The results of several classification experiments on variously oriented samples of the visible wavelength targets will be presented.
In this contribution, we propose an optoelectronic hit/miss morphological transform for real-time quality control by image moire analysis, that integrates a VanderLugt optical correlator and a digital signal processor...
详细信息
ISBN:
(纸本)0819425214
In this contribution, we propose an optoelectronic hit/miss morphological transform for real-time quality control by image moire analysis, that integrates a VanderLugt optical correlator and a digital signal processor associated to a vector co-processor. The procedure for real-time defect detection is a three stage process. The first step is to enhance the moire image, using the wavelet decomposition and a multiresolution approach. The second step is to automatically segment the enhanced moire image, using the moment preserving thresholding algorithm. The third step is to apply the morphological hit/mass transform to directly recognize moire images that correspond to defectuous objects. This new procedure of defect detection by global analysis of moire image data has been compared to another new technique that is based on multidimensional supervised classification of optical correlations between the test object moire image and reference moire images.
Most current standards for video compression employ block matching motion compensation to reduce temporal redundancy, although they do not specify how the block matching is to be implemented. Exhaustive search gives t...
详细信息
ISBN:
(纸本)085296692X
Most current standards for video compression employ block matching motion compensation to reduce temporal redundancy, although they do not specify how the block matching is to be implemented. Exhaustive search gives the optimal compressed video signal quality for block matching but at the expense of high computational complexity. This computation can be decreased by well-known reduced search techniques, such as log search, but this can be at the expense of reduced signal to noise ratio. This paper presents two techniques which maintain signal quality close to that of full search but with significant savings in computation. The first technique employs multiple searches focused on the predominant small motion vectors combined with pixel decimation to reduce comparisons. The second dynamically adjusts the search window size to avoid wasting search effort on areas where it is considered unlikely that the optimal motion vector will be found.
A general technique based on the use of structure function is used to synthesize and characterize anisotropic nonstationary fields with stationary increments. The classical fBm model is found to be a particular case o...
详细信息
ISBN:
(纸本)085296692X
A general technique based on the use of structure function is used to synthesize and characterize anisotropic nonstationary fields with stationary increments. The classical fBm model is found to be a particular case of this class of processes. The diversity of the textures generated using this method is connected to the admissible solutions for the structure function. The fundamental definitions concerning fields with stationary increments and the conditions satisfying the structure function are presented.
The fractal method offers high compression ratio, good image quality, and resolution independence of the decoded image. Its disadvantage is the long encoding time due to the repeated search of the domain pool. A techn...
详细信息
ISBN:
(纸本)085296692X
The fractal method offers high compression ratio, good image quality, and resolution independence of the decoded image. Its disadvantage is the long encoding time due to the repeated search of the domain pool. A technique is developed for speeding up the fractal encoding based on a combination of the tree method and the traditional block classification. It can lead up to 80% speed-up gain against the class-based search method.
暂无评论