Batik, as a cultural heritage from Indonesia, has a lot of motifs based on certain patterns. This paper discusses feature extraction methods for the recognition of batik motifs in digital images. In this study, the us...
详细信息
Batik, as a cultural heritage from Indonesia, has a lot of motifs based on certain patterns. This paper discusses feature extraction methods for the recognition of batik motifs in digital images. In this study, the use of several feature extraction methods have been compared in terms of their performance with several scenarios for testing level accuracy. The methods include Gray Level Co-occurrence Matrices (GLCM), Canny Edge Detection, and Gabor filters. The experimental results show that the use of GLCM features has performed the best with a classification accuracy reaching 80%.
This paper presents a paddy growth stages classification using MODIS remote sensing images with support vector machines (SVMs). We collected the paddy growth stages data samples from a series of MODIS mages acquired f...
详细信息
This paper presents a paddy growth stages classification using MODIS remote sensing images with support vector machines (SVMs). We collected the paddy growth stages data samples from a series of MODIS mages acquired from March to July 2012 along paddy field area only. The data are collected based on growth stages phenology of paddy using spectral profile which consists of at least 9 classes for growth stages and 2 classes for dominated soil and cloud. We apply SVMs to build a binary classifier for each class with one against all strategy of multiclass approach. One important issue needed to address is unbalanced prior probability that should be solved by each SVM. In this study, we evaluate the effectiveness of balanced branches strategy that is applied to one against all SVMs learning. Our results shows that the balanced branches strategy does improves in average around 10% classification accuracy during training and validation, and in average around 50% during testing.
In many application fields, an appropriate high-quality fast image upsampling method is required. Although many interpolation-based upsampling methods have been proposed, the quality of result images is not satisfacto...
详细信息
In many application fields, an appropriate high-quality fast image upsampling method is required. Although many interpolation-based upsampling methods have been proposed, the quality of result images is not satisfactory. Some of them are very fast, but produce poor quality images, the others can produce high quality images, but the methods in them are slow. In our paper, we proposed a fast statistical image upsampling method based on CUDA, it can obtain high quality images based on reducing the input resolution-grids dependency artifacts. Thus, we can rebuild low resolution images' sharp edges fast and get high-quality upsampled images in real time. We have applied this method in the multi-resolution texture generation of large scale terrain rendering. Experiments prove that our method can receive ideal effects in real time.
In this paper, we propose a novel approach for image completion with automatic structure propagation. This method integrates two stages: Firstly, it extends the salient structure lines from the known regions to the un...
详细信息
In this paper, we propose a novel approach for image completion with automatic structure propagation. This method integrates two stages: Firstly, it extends the salient structure lines from the known regions to the unknown by following a local self-similarity assumption on natural images. Then guided by the structure information, it restores the missing region by patch-based texture synthesis. Experiment results demonstrate a better effect of our method than that of the previous patch-based texture synthesis image completion algorithm.
The reliable estimation of system state in multi-sensor uncertainty is always the hot and knotty issue of nonlinear filtering theory. Aiming to the reasonable utilization of measurement information, a novel multi-sens...
详细信息
The license plate location technique is an important imageprocessing step in license plate recognition system. Vehicle license plates are distinguished from backgrounds using features proposed in existing literatures...
详细信息
The license plate location technique is an important imageprocessing step in license plate recognition system. Vehicle license plates are distinguished from backgrounds using features proposed in existing literatures. However, the effect of location is quite affected by feature selection. In this paper, we propose a method of precise license plate location fusing salient features. The method is mainly divided into three steps. First, candidate license plate regions are detected using improved Harris corner feature with much less time than traditional method. Then, candidates are sifted to only retain license plates based on two salient features named color combination and mean difference which are first proposed in this paper. Finally, the license plates are located precisely according to the projection feature. In experiment, the proposed algorithm was tested with 1942 real images captured in different environment and the license plates are successfully located as 97.6% in average with only 109ms. The experiment results demonstrates the effectiveness and efficient of our algorithm.
Character information is hard to detect in billet scene images by CCD camera. In this paper, we present a method for detection of billet characters from measurements of recursive segmented image. This recursive segmen...
详细信息
Character information is hard to detect in billet scene images by CCD camera. In this paper, we present a method for detection of billet characters from measurements of recursive segmented image. This recursive segmented method can be used in a wide variety of billet scenes. According to high temperature and complex scene in the rolling line, we use an effective clustering and projection characteristics to determine the terminal condition of recursive segmentation. Then we can label character candidate regions in turn by this effective characteristics, and select the regions we want to achieve. The experiments show that this method makes full use of the characteristics of region and clustering. It can improve the quality of detection, and the detection result meets the need of practical application.
Tracking the same person across multiple cameras is an important task in multi-camera systems. It is also desirable to re-identify the individuals who have been previously seen with a single-camera. This paper address...
详细信息
Tracking the same person across multiple cameras is an important task in multi-camera systems. It is also desirable to re-identify the individuals who have been previously seen with a single-camera. This paper addresses this problem by the re-identification of the same individual in two different datasets, which are both challenging situations from video surveillance system. In this paper, local descriptors are introduced for image description, and support vector machines are employed for high classification performance and so an efficient Bag of Features approach for image presentation. In this way, robustness against low resolution, occlusion and pose, viewpoint and illumination changes is achieved in a very fast way. We get promising results from the evaluation with situations where a number of individuals vary continuously from a multi-camera system.
An accuracy assessment method that integrates segmentation and classification accuracy is proposed to meet the requirements of object-based image analysis. Segmentation errors are measured by establishing the relation...
详细信息
ISBN:
(纸本)9781467301732
An accuracy assessment method that integrates segmentation and classification accuracy is proposed to meet the requirements of object-based image analysis. Segmentation errors are measured by establishing the relationship between pixels and their corresponding segments according to the overlaps of segments and reference polygons. Then, two improved confusion matrices that take the segmentation errors into consideration are used: one for pixel-level classification results, and the other for object-level classification results. A final accuracy assessment combines the statistics of these two confusion matrices. The proposed method can be applied to segmentation scale selection in the hierarchical interpretation system. An experiment on a SPOT5 image demonstrates the effectiveness of this method for segmentation scale selection, which can guide the fusion of objects of different scales to obtain a higher accuracy.
Character information is hard to detect in billet scene images by CCD camera. In this paper, we present a method for detection of billet characters from measurements of recursive segmented image. This recursive segmen...
详细信息
Character information is hard to detect in billet scene images by CCD camera. In this paper, we present a method for detection of billet characters from measurements of recursive segmented image. This recursive segmented method can be used in a wide variety of billet scenes. According to high temperature and complex scene in the rolling line, we use an effective clustering and projection characteristics to determine the terminal condition of recursive segmentation. Then we can label character candidate regions in turn by this effective characteristics, and select the regions we want to achieve. The experiments show that this method makes full use of the characteristics of region and clustering. It can improve the quality of detection, and the detection result meets the need of practical application.
暂无评论