摘要 :
Feature extraction plays an important role in face recognition. Based on local binary patterns (LBP), we propose a novel face representation method which obtains histograms of semantic pixel sets based LBP (spsLBP) with a robust c...
展开
Feature extraction plays an important role in face recognition. Based on local binary patterns (LBP), we propose a novel face representation method which obtains histograms of semantic pixel sets based LBP (spsLBP) with a robust code voting (rcv). By clustering according the semantic pixel relations before the histogram estimation, the spsLBP makes better use of the spatial information over the original LBP. In this paper, we use a simple rule to use the semantic information. We cluster by the pixel intensity values, which is also invariant to monotonic grayscale changes, and it is in particular very useful when there are occlusions and expression variations on face images. Besides, the proposed representation adopts a new code voting strategy for LBP histogram computation, which makes it more robust. The proposed method is evaluated on three widely used face recognition databases: AR, FERET and LFW. Experimental results show that the proposed method can outperform the original uniform LBP and its extensions.
收起
摘要 :
As there is a need for interpretable classification models in many application domains, symbolic, interpretable classification models have been studied for many years in the literature. Rule-based models are an important class of ...
展开
As there is a need for interpretable classification models in many application domains, symbolic, interpretable classification models have been studied for many years in the literature. Rule-based models are an important class of such models. However, most of the common algorithms for learning rule-based models rely on heuristic search strategies developed for specific rule-learning settings. These search strategies are very different from those used in neural forms of machine learning, where gradient-based approaches are used. Attempting to combine neural and symbolic machine learning, recent studies have therefore explored gradient-based rule learning using neural network architectures. These new proposals make it possible to apply approaches for learning neural networks to rule learning. However, these past studies focus on unordered rule sets for classification tasks, while many common rule-learning algorithms learn rule sets with an order. In this work, we propose RL-Net, an approach for learning ordered rule lists based on neural networks. We demonstrate that the performance we obtain on classification tasks is similar to the state-of-the-art algorithms for rule learning in binary and multi-class classification settings. Moreover, we show that our model can easily be adapted to multi-label learning tasks.
收起
摘要 :
Local Binary Patterns is one of the most effective approaches for pattern recognition in general and face recognition in particular. There have been many studies on improving this method such as changing the input values or using ...
展开
Local Binary Patterns is one of the most effective approaches for pattern recognition in general and face recognition in particular. There have been many studies on improving this method such as changing the input values or using another kind of histogram. Although weight set is also an important key leading to the success of this method, it does not seem to get much attention. A majority of LBP-based approaches are still using the weight set of Ahonen et al.'s study, one of the first researches applying LBP to face recognition. In this study, we introduce a powerful algorithm named Heuristic Weight Search, which finds a suitable weight set for not only LBP-based approaches but also other methods using weight set to improve performance. Experiments on the FERET database prove an ability of HWS thanks to their higher accuracy than original methods.
收起
摘要 :
Local Binary Patterns is one of the most effective approaches for pattern recognition in general and face recognition in particular. There have been many studies on improving this method such as changing the input values or using ...
展开
Local Binary Patterns is one of the most effective approaches for pattern recognition in general and face recognition in particular. There have been many studies on improving this method such as changing the input values or using another kind of histogram. Although weight set is also an important key leading to the success of this method, it does not seem to get much attention. A majority of LBP-based approaches are still using the weight set of Ahonen et al.'s study, one of the first researches applying LBP to face recognition. In this study, we introduce a powerful algorithm named Heuristic Weight Search, which finds a suitable weight set for not only LBP-based approaches but also other methods using weight set to improve performance. Experiments on the FERET database prove an ability of HWS thanks to their higher accuracy than original methods.
收起
摘要 :
Pattern Matching is useful for recognizing character in a digital image. Optical Character Recognition (OCR) is one such technique which reads character from a digital image and recognizes them. Line segmentation is initially used...
展开
Pattern Matching is useful for recognizing character in a digital image. Optical Character Recognition (OCR) is one such technique which reads character from a digital image and recognizes them. Line segmentation is initially used for identifying character in an image and later refined by morphological operations like binarization, erosion, thinning etc. The work discusses a recognition technique that defines a set of morphological operators based on its orientation in a character. These operators are further categorized into groups having similar shape but different orientation for efficient utilization of memory. Finally the characters are recognized in accordance with the occurrence of frequency in hierarchy of significant pattern of those morphological operators and by comparing them with the existing database of each character.
收起
摘要 :
Pattern Matching is useful for recognizing character in a digital image. Optical Character Recognition (OCR) is one such technique which reads character from a digital image and recognizes them. Line segmentation is initially used...
展开
Pattern Matching is useful for recognizing character in a digital image. Optical Character Recognition (OCR) is one such technique which reads character from a digital image and recognizes them. Line segmentation is initially used for identifying character in an image and later refined by morphological operations like binarization, erosion, thinning etc. The work discusses a recognition technique that defines a set of morphological operators based on its orientation in a character. These operators are further categorized into groups having similar shape but different orientation for efficient utilization of memory. Finally the characters are recognize d in accordance with the occurrence of frequency in hierarchy of significant pattern of those morphological operators and by comparing them with the existing database of each character.
收起
摘要 :
Pattern Matching is useful for recognizing character in a digital image. Optical Character Recognition (OCR) is one such technique which reads character from a digital image and recognizes them. Line segmentation is initially used...
展开
Pattern Matching is useful for recognizing character in a digital image. Optical Character Recognition (OCR) is one such technique which reads character from a digital image and recognizes them. Line segmentation is initially used for identifying character in an image and later refined by morphological operations like binarization, erosion, thinning etc. The work discusses a recognition technique that defines a set of morphological operators based on its orientation in a character. These operators are further categorized into groups having similar shape but different orientation for efficient utilization of memory. Finally the characters are recognize d in accordance with the occurrence of frequency in hierarchy of significant pattern of those morphological operators and by comparing them with the existing database of each character.
收起
摘要 :
For many years, one of the problems in pattern recognition is classification. There are many methods that deal with this type of problem. The data sets are sometimes in the binary form (real number) and represented by vectors of b...
展开
For many years, one of the problems in pattern recognition is classification. There are many methods that deal with this type of problem. The data sets are sometimes in the binary form (real number) and represented by vectors of binary numbers (real numbers) although there are uncertainties in the data, e.g., data collected in management questionnaires. In this paper, we developed a linguistic K-nearest prototype algorithm with vectors of fuzzy numbers as inputs. This algorithm is based on the extension principle and the decomposition theorem. We apply this algorithm to linguistic vectors derived from a set of thirty-nine subjects answering questions about students' satisfaction with communication to their university.
收起
摘要 :
A novel method of texture image segmentation is proposed, which has three advantages compared to other active contours. First, by combining the gray levels of pixels with texture information of an image,this method can be used for...
展开
A novel method of texture image segmentation is proposed, which has three advantages compared to other active contours. First, by combining the gray levels of pixels with texture information of an image,this method can be used for segmentation of a texture image or a non-texture image. Second, the method has low computation complexity because local binary pattern (LBP) is employed to extract texture features.And last, the proposed algorithm can avoid the additional computation problem without re-initialization of signal distance function repeatedly.The segmentation tests show that the proposed segmentation method is efficient, accurate, fast and robust.
收起
摘要 :
A novel method of texture image segmentation is proposed, which has three advantages compared to other active contours. First, by combining the gray levels of pixels with texture information of an image,this method can be used for...
展开
A novel method of texture image segmentation is proposed, which has three advantages compared to other active contours. First, by combining the gray levels of pixels with texture information of an image,this method can be used for segmentation of a texture image or a non-texture image. Second, the method has low computation complexity because local binary pattern (LBP) is employed to extract texture features.And last, the proposed algorithm can avoid the additional computation problem without re-initialization of signal distance function repeatedly.The segmentation tests show that the proposed segmentation method is efficient, accurate, fast and robust.
收起