In computervision research field shape matching plays a central role. A typical approach is based on the analysis of contour points of the objects. In many cases this task is faced by taking into account a contour su...
详细信息
In computervision research field shape matching plays a central role. A typical approach is based on the analysis of contour points of the objects. In many cases this task is faced by taking into account a contour subset made up by dominant points. This paper is based on a shape matching technique described in [1] by using dominant points as contour points set. In order to obtain better results we adopt modified shape descriptors (shape contexts) by getting rotational invariance. Experimental results demonstrate the accuracy improvement and the computational time reduction.
Point pattern matching is an important topic in computervision and patternrecognition, and finds many applications such as image registration, motion detection, object tracking and pose estimation. In this paper, we...
详细信息
Point pattern matching is an important topic in computervision and patternrecognition, and finds many applications such as image registration, motion detection, object tracking and pose estimation. In this paper, we propose an efficient algorithm for determining correspondence between two planar point sets under transform of translation, rotation and scale. This algorithm randomly selects some points of a set and extracts their neighbor points. It views the selected points and their neighbor points as local point patterns, and finds the local matched patterns in the other set. Point pattern matching is finally achieved by counting the unique point number of those local matched point patterns with the same transform parameters. Many experiments are conducted to validate efficiency of the proposed algorithm. Running time comparisons with a well-known point pattern matching algorithm are also done and the results show that the proposed algorithm is faster than the compared algorithm. (C) 2014 Elsevier Inc. All rights reserved.
Crowd counting which aims at obtaining the number of people within a scene is an important computervision task. While most previous methods try to count people within one frame, this paper addresses this problem usin...
详细信息
Crowd counting which aims at obtaining the number of people within a scene is an important computervision task. While most previous methods try to count people within one frame, this paper addresses this problem using the detection flow which is defined as a set of object detection responses along the temporal video sequence. We argue that counting based on detection flow provides a better way to estimate the crowd size with following merits: 1) it can greatly alleviate the common weakness of an object detector including miss detection and false alarms; 2) it is robust to temporal object occlusions and noises; 3) it is more competent to give specific descriptions of the crowd, e.g. crowd moving directions and target locations. Experiment results on PETS 2009 dataset demonstrate the potential of this method.
This paper presents a new solution to the automatic localization and recognition of natural digits embossed on structured and rough surfaces. A new concept of combining periodic (e.g. FFT, DCT) with polynomial moments...
详细信息
This paper presents a new solution to the automatic localization and recognition of natural digits embossed on structured and rough surfaces. A new concept of combining periodic (e.g. FFT, DCT) with polynomial moments (e.g. Tchebichef moments) is shown that the new method can eliminate some of the problems associated with the Gibbs phenomena. QR decomposition is used to obtain a unitary basis, minimizing the numerical effort when modeling surfaces. The combined moment method is used to generate a smoothed global model of the surface structure. Following segmentation of the individual digits, patternrecognition based on moment function such as Hu’s moment invariants, Zernike moments, and Orthogonal Fourier Mellin Moments was used to recognize the stamped number after analyzing 3D surface data of the steel block. A prototype system of the laser-scanning instrument was implemented and test measurements were performed in a production line. The results showed the correct functionality of the method.
Single image haze removal (or de-haze) using dark channel prior model is effective when there exists a dark channel within the image. So for the images which do not meet the dark channel prior, the haze removal result...
详细信息
Single image haze removal (or de-haze) using dark channel prior model is effective when there exists a dark channel within the image. So for the images which do not meet the dark channel prior, the haze removal result may appear light pollution, cross-color. In this paper, we propose an algorithm to judge whether a single image meets the dark channel prior, and for the image that fails to meet the prior, it eliminates the cross-color influence in the image after de-hazing. Experimental results show that ours can determine whether the image meets dark channel prior, and has the better de-haze effect.
The proposed scheme is an approach which can be used to improve the performance of traditional image segmentation systems. The scheme is based on a framework that employs the output of an existing image segmentation p...
详细信息
The proposed scheme is an approach which can be used to improve the performance of traditional image segmentation systems. The scheme is based on a framework that employs the output of an existing image segmentation process together with hierarchical clustering using an information theoretic similarity measure. Experimental results clearly show that when the scheme operates in conjunction with a state of the art image segmentation algorithm, it yields significantly superior performance over a wide spectrum of natural images. These results are based on informal subjective evaluation tests as well as on objective measurements obtained from processing the Berkeley BSDS 300 image dataset.
In this study, main goal is to estimate scene illuminant chromaticity values for each of the three color channels (red, green, blue). It is found that analysis of the pixel values, which are expressed by dichromatic r...
详细信息
In this study, main goal is to estimate scene illuminant chromaticity values for each of the three color channels (red, green, blue). It is found that analysis of the pixel values, which are expressed by dichromatic reflection and inverse-intensity chromaticity space definitions, with appropriate statistical methods gives aproximate values for light source chromaticity within an acceptable error range. Single image of a scene is sufficient for calculations. We saw that chromaticity values of light sources with different colors can be approximately calculated.
In order to reduce the detection time of Escherichia coli(E. coli) on fresh, minimally processed vegetables, a new close to real-time method based on artificial intelligence technology is presented. The ability of thr...
详细信息
In order to reduce the detection time of Escherichia coli(E. coli) on fresh, minimally processed vegetables, a new close to real-time method based on artificial intelligence technology is presented. The ability of three-layer multilayer perceptron neural network(MLP) models to predict E. coli contaminated condition using feature parameters extracted from E. coli images was compared with multivariate statistical model(discriminant analysis, DA) with respect to the identification accuracy of E. coli cells. The performance of MLP was 10% better than that from the DA. Comparisons of the counts of E. coli by rapid method and by traditional method(multiple-tube fermentation technique) closed correlated(R2=0.9867). Besides the high accuracy, the total detection time was within 1 h and the detection limit was 10 cells /ml. This is a clear improvement compared with the 6 days taken to obtain results from the traditional method.
The problem of calibrating paracatadioptric camera from sphere images is still open. In this paper, we propose a calibration method for paracatadioptric camera based on spheres. We notice that, under central catadiopt...
详细信息
The problem of calibrating paracatadioptric camera from sphere images is still open. In this paper, we propose a calibration method for paracatadioptric camera based on spheres. We notice that, under central catadioptric camera, a sphere is projected to two conies on the image plane, which can also be seen as the projections of two parallel circles on the viewing sphere by a virtue camera. These two conies are called a pair of antipodal sphere images. Firstly, we study properties of K(K >; 3) pairs of antipodal sphere images under paracatadioptric camera. Then, they can be estimated using these properties. Finally, paracatadioptric camera can be calibrated by three pairs of antipodal sphere images or more. The method only requires the projected contour of parabolic mirror is visible on the image plane in one view. Experimental results on both simulated and real image data have demonstrated the effectiveness of our method.
Wii Remote is the standard controller of the game console Nintendo Wii®. Despite its low cost, it has a very high performance and high resolution infrared camera. It also has a built-in chip for tracking up to 4 ...
详细信息
Wii Remote is the standard controller of the game console Nintendo Wii®. Despite its low cost, it has a very high performance and high resolution infrared camera. It also has a built-in chip for tracking up to 4 points from the infrared images. With these properties, the Wii Remote has attracted several attempts to calibrate and use them for metric 3D information. Thanks to Wii Remote's 4 point tracking hardware, people were able to calibrate Wii Remotes using carefully designed square patterns with infrared LEDs at each corner. However, it is not easy for average Wii Remote users to implement such calibration tools. In this paper, we give an overview of the current Wii Remote calibration implementations and present a novel method to fully calibrate multiple Wii Remotes using only two points, such as the LEDs of the Sensor Bar that comes with the Wii console. For the 3D reconstruction, only two Wii Remotes and a PC is needed. We also demonstrate the usability of a full 3D tracker with a multiplayer game that receives inputs from up to 4 users with only two Wii Remotes.
暂无评论