作者:
Erdal YenialpHabil KalkanVision
Image Processing and Pattern Recognition Laboratory (VIPLAB) Bilgisayar Mühendisliği Bölümü Süleyman Demirel Üniversitesi Isparta Turkey
Segmentation algorithms are widely used in imageprocessing. These methods have different complexity values and the choice of reasonable methods decreases on large images. Especially on the medical images with large s...
详细信息
Segmentation algorithms are widely used in imageprocessing. These methods have different complexity values and the choice of reasonable methods decreases on large images. Especially on the medical images with large size, it may take days to perform segmentation in some methods. However, parallel implementation may eliminate the drawback of these algorithms to some extent. In this study, we propose to implement segmentation algorithms in parallel using Graphical processing Unit. Using the proposed implementation, the computation time of the K-centers, K-means and DBSCAN algorithms were decreases 87, 642 and 2 times, respectively.
Aiming at the disadvantages of the traditional off-line vector-based learning algorithm, this paper proposes a kind of Incremental Tensor Principal Component Analysis (ITPCA) algorithm. It represents an image as a ten...
详细信息
Lines provide important information in images and line detection is crucial in many applications. Many line features can be used to detect line position while line width (i.e., thickness) is a more structured, higher-...
详细信息
ISBN:
(纸本)9781479900145
Lines provide important information in images and line detection is crucial in many applications. Many line features can be used to detect line position while line width (i.e., thickness) is a more structured, higher-level feature compared to edge or other line features. Every point of the wide line structure has its own width in spite of the structure's asymmetry. In this paper, we use the parallel edges to get the line orientation and width. We do not need to know the actual width of the line, but rather recover it to get the region of interest (ROI). Then use a modified OTSU obtaining proper region threshold T to segment the wide line structures. By fusing feature of width and gray, a sequence of tests has been conducted on a variety of image samples obtained from simple natural scene and our experimental results demonstrate the practical and robust of the proposed method.
In order to settle incremental learning and preserve the space information of images, this paper proposes an incremental tensor discriminant analysis for facial image detection. The proposed algorithm employs tensor r...
详细信息
Human action recognition is a hot topic in computer vision field. Various applicable approaches have been proposed to recognize different types of actions. However, the recognition performance deteriorates rapidly whe...
详细信息
ISBN:
(纸本)9781479923427
Human action recognition is a hot topic in computer vision field. Various applicable approaches have been proposed to recognize different types of actions. However, the recognition performance deteriorates rapidly when the viewpoint changes. Traditional approaches aim to address the problem by inductive transfer learning, in which target-view samples are manually labeled. In this paper, we present a novel approach for cross-view action recognition based on transduc-tive transfer learning. We address the problem by transferring instances across views. In our settings, both labels of examples from the target view and the corresponding relation between examples from pairwise views are dispensable. Experimental results on the IXMAS multi-view data set demonstrate the effectiveness of our approach, and are comparable to the state of the art.
Object classification in traffic scene surveillance has attracted much attention recent years. Traditional classification methods need lots of labeled samples to build a satisfying classifier. However, the acquisition...
详细信息
ISBN:
(纸本)9781479923427
Object classification in traffic scene surveillance has attracted much attention recent years. Traditional classification methods need lots of labeled samples to build a satisfying classifier. However, the acquisition of the labeled samples may cost lots of time and human labor. In this paper, we propose an label-propagation based semi-supervised learning method which uses the information of both labeled and unlabeled samples. Experiment results show that our method outperforms the traditional methods both in accuracy and robustness.
pattern matching is a fundamental application text retrieval, string query, biological sequence analysis, etc. Therefore, the effective algorithm performing this kind of matching is in great need. In this paper, the w...
详细信息
pattern matching is a fundamental application text retrieval, string query, biological sequence analysis, etc. Therefore, the effective algorithm performing this kind of matching is in great need. In this paper, the wildcard is defines to match any one character in a sequence. Multiple wildcards form a gap. The length of a flexible gap is arbitrary. We design CLPM algorithm by use of cross list index structure to realize pattern matching with flexible wildcard gaps. The preprocessing algorithm is designed to initialize cross list so as to reduce searching space. In CLPM algorithm, the effective intervals is defined and computed based on the start positions of each sub pattern in each string, which help to obtain matching result set. Moreover, the approximate pattern matching is converted to short extract pattern matching. The contrast experiments are done based on DBLP tile data set. The results show that CLMP algorithm has better performance in the same fields.
pattern Mining is a popular issue in biological sequence analysis. With the introduction of wildcard gaps, more interesting patterns can be mined. In this paper, we propose a new definition related to pattern frequenc...
详细信息
pattern Mining is a popular issue in biological sequence analysis. With the introduction of wildcard gaps, more interesting patterns can be mined. In this paper, we propose a new definition related to pattern frequency, under which the Apriori property holds. We define a pattern mining problem called Ming top-K Frequent patterns (MFP), where gaps are mined instead of specified. Compared with existing problems, MFP does not require any domain knowledge of the user. However, theoretical analysis and experimental results show that MFP favors inflexible patterns. We then define another problem where the flexibility threshold of each gap is specified by the user. The problem is called Mining top-K Frequent and Flexible patterns (MF 2 P). We develop algorithm with polynomial complexities for both problems. patterns can grow from both sides. Some interesting biological patterns mined by our algorithms are discussed.
image segmentation is the basis of imageprocessing and image analysis. However, there are no common method that can be used in natural images, and present methods fail to explain understandings of human's visual ...
详细信息
Raisins grade identification in China still relies on photoelectric sorting and manual separation, also, the function of management system for the production, processing, and sales of raisin is traditional and simple....
详细信息
暂无评论