Narrow Results

Skin detection plays an important role in applications such as face detection and tracking, person detection and pornography detection. While previous studies focus on pixel-based skin color detection techniques that individually classify each pixel as skin color or non-skin color, this study presents a region-based algorithm for detecting skin color. The proposed algorithm uses a special region, called key skin region, as the basis to classify skin color. A performance comparison with conventional skin classifiers, including the Bayesian classifier, the unimodal Gaussian classifier and the Gaussian mixture classifier, is made in this study. Experimental results show that the proposed algorithm outperforms other tested skin classifiers. Furthermore, the skin regions detected by the proposed algorithm, especially facial regions, are nearly complete with no hollow holes in these regions. This property can simplify the complexity of implementing applications that use skin color as their basis, such as face detection and face tracking.

Face recognition plays a vital role in authentication, monitoring, indexing, access control and other surveillance applications. Much research on face recognition with various feature based approaches using global or local features employing a number of similarity measurement techniques have been done earlier. Feature based approaches using global features can effectively preserve only the Euclidean structure of face space, that suffer from lack of local features which may play a major role in some applications. On the other hand, wtih local features only the face subspace that best detects the essential face manifold structure is obtained and it also suffers loss in global features which may also be important in some other applications. Measuring similarity or distance between two feature vectors is also a key step for any pattern matching application. In this work, a new combined approach for recognizing faces that integrates the advantages of the global feature extraction technique by Linear Discriminant Analysis (LDA) and the local feature extraction technique by Locality Preserving Projections (LPP) with correlation based similarity measurement technique has been discussed. This has been validated by performing various experiments and by making a fair comparison with conventional methods.

A novel approach for feature extraction of fingerprint matching is proposed by using two-dimensional (2D) rotated wavelet filters (RWF). 2D RWF are used to capture the characterization of diagonally oriented information present in fingerprint image. Proposed method extracts the significant information from small area of fingerprint image. Experimental results conducted on standard database of Bologna University and FVC2002 indicate that the proposed method improves the genuine acceptance rate (GAR) from 92.14% to 96.12% and reduces false acceptance rate (FAR) from 25.2% to 21.2% on Bologna University database and it reduces FAR from 36.71% to 22.79% on FVC2002 database compared with discrete wavelet transform-based approach.

With the rapid development of remote sensing technology, searching the similar image is a challenge for hyperspectral remote sensing image processing. Meanwhile, the dramatic growth in the amount of hyperspectral remote sensing data has stimulated considerable research on content-based image retrieval (CBIR) in the field of remote sensing technology. Although many CBIR systems have been developed, few studies focused on the hyperspectral remote sensing images. A CBIR system for hyperspectral remote sensing image using endmember extraction is proposed in this paper. The main contributions of our method are that: (1) the endmembers as the spectral features are extracted from hyperspectral remote sensing image by improved automatic pixel purity index (APPI) algorithm; (2) the spectral information divergence and spectral angle match (SID–SAM) mixed measure method is utilized as a similarity measurement between hyperspectral remote sensing images. At last, the images are ranked with descending and the top-$M$ retrieved images are returned. The experimental results on NASA datasets show that our system can yield a superior performance.

This paper presents a new Chinese character similarity measurement method based on the ring projection algorithm and distance transform. The ring projection algorithm is used to transform a character image with two independent variables into a function of one independent variable in the ring projection space. This representation of character in the ring projection space has been proved to be in orientation and scale invariant. However, this representation will be distorted nonlinearly in the presence of noise. Therefore, common linear metrics such as Euclidean distance, cannot be applied to measure distance. To solve the nonlinear distortion problem, distance transform is proposed as a nonlinear metric. The similarity measurement is performed using the distance transformed image in the ring projection space. A number of Chinese characters are selected to evaluate the capability of the proposed measurement scheme and the results are encouraging.

Concept development is the first and most knowledge-intensive step in the development process of manufacturing technology. Its core is to find a solid scientific foundation for the manufacturing requirements in order to propose a feasible manufacturing technology concept. However, the lack of formal methods and finiteness of personal knowledge result in high randomness and low efficiency of this step. This paper presents a formal design knowledge recommendation method for manufacturing technology concept development by calculating the morphological similarity between manufacturing requirements and multi-domain effect knowledge. In this method, the morphological matrix of general manufacturing technology is constructed first as a template. Then, manufacturing requirements and multi-domain effect knowledge are both expressed as matrices based on this template. Finally, through quantitatively calculating the normalized weighted Euclidean distance between manufacturing requirements and multi-domain effect knowledge, suitable effects of knowledge which are from different domains and have the domain-highest similarity are recommended as the initial design foundation for the concept of new manufacturing technology (NMT). A software system has been developed and a concept development case of composite machining technology was provided to validate this method. The result shows that the proposed approach can reduce the randomness and increase the efficiency of manufacturing technology concept development.

This paper presents a new Chinese character similarity measurement method based on the ring projection algorithm and distance transform. The ring projection algorithm is used to transform a character image with two independent variables into a function of one independent variable in the ring projection space. This representation of character in the ring projection space has been proved to be in orientation and scale invariant. However, this representation will be distorted nonlinearly in the presence of noise. Therefore, common linear metrics such as Euclidean distance, cannot be applied to measure distance. To solve the nonlinear distortion problem, distance transform is proposed as a nonlinear metric. The similarity measurement is performed using the distance transformed image in the ring projection space. A number of Chinese characters are selected to evaluate the capability of the proposed measurement scheme and the results are encouraging.