Papers by Keyword: Edge Detection

Abstract: Detecting human existence in video streams is a fundamental task in many video processing applications. In this paper, a novel procedure is produced to model, analyze and recognize human motions (jogging and walking in dark environment) in video streams. There are four major areas that are related in this project for human motion analysis: (1) developing human body structure based on human skeleton model, (2) tracking and data collecting human motion with side view, (3) recognizing human activities from image sequences, and (4) image processing technique using edge detection and vectors angle calculation. All algorithms are developed using MATLAB software. Segmentation is developed to reduce the amount of data and filters out the useless information. Two methods are proposed for angle calculation and activities classification. Results showed that angle between 153.76°-180° for method 1 and 49.64°-92.86° for method 2 is classified as walking while jogging is 95.17°-138.72° for method 1 and 22.62°-56.31° for method 2.

Abstract: Edge detection is the most basic arithmetic of the local information of image change detection, image edge detection can help people better describe or identify image, is abasic technique in image processing. Edge detection is mainly to find out the important goal of the salient points on the boundary of the image,because the traditional edge detection operators have the same direction, so easily missed in the detection process.In this paper, the solitary wave algorithm for image edge detection, this algorithm can provide both the direction and size of the image edge, a certain extent, can also solve the missing edges, it is possible to extract more accurate image information.

Abstract: In industrial production, the machine vision save a lot of manpower and material resources. And it has become an indispensable part of quality control solutions in automobile and parts manufacturing. In order to detect the integrity of automatic assembly for lamp parts, we first use the shape feature to determine whether the part shape is right through egde detection and template matching algorithms. And if the shape is right, we can further detect the color feature which are different in different sales regions using the color moments. As a result, the missing and mistake packing can be detected.

Abstract: Hough transform as an effective graphics target detection method can detect straight lines, circles, ellipses, parabolas and many other analytical graphics. The discretization of space, as well as the calculation of the process make Hough transform have some limitations, such as poor detection results because of high-intensity noise, a large amount of calculation, large demand of storage resources and so on. This paper analyzes the Hough Transform voting process and points out that the accumulation with 1 in the method is unreasonable. The paper proposed a Hough transform based on template matching via the modification of the definition of the traditional method. In this method, each parameter unit identifies a template in image space. The feature points according with the conditions can be searched by the template actively. The method takes the number of feature points as the value of parameter unit and takes the record of the coordinates of line segment endpoints. So line segments can be detected and storage resources can be saved.

Abstract: The edge detection is very important in the early stage of the visual processing, According to deal with the local characteristics of image information, First,this paper apply GAP predictor templates prediction error of the image. Then using improved genetic algorithm combined with OTSU method, calculate the threshold. According the threshold, the error image edge was classified , and image edge was extracted. Experiments show that this paper algorithm is better results than the traditional edge detection, It has the advantage of edge location accurate, less false edge, and so image edge detection is an effective supplement.

Abstract: This paper proposes dynamic threshold control algorithm by Co-evolutionary implementation base on a multidirectional gradient edge detection predictor (MGEDP) template. First, the image cut into four equal parts, these parts will be calculate simultaneously the error values by MGEDP predictor template, in the four sub images using, dynamic threshold evolution; Use these feedback values, to get the error image, and to calculate the threshold values by collaborative computing OTSU threshold, and so on, to classify the edges of error image, The result was convinced that this model could not only accelerated the image processing by Co-evolutionary implementation but get the more details, clearer edges, and better visual sense image.

Abstract: Sobel, Roberts operator is derived based on the differential. As a result of the template and fixed threshold value, it lacks adaptability. Median filter on images of the collected cardiac seeds maturity, use split clustering algorithm on cardiac seeds maturity for the first image gradient value clustering, condensed cluster on the result of the first split the clustering results, then secondly division cluster, and finally work out the image edge based on the second clustering results and realize on the FPGA implementation. At last the method is Applied in Andriod Platform. The experimental results show that it is more delicate to use the hierarchical clustering algorithm to detect the edge and it has stronger ability to suppress noise.

Abstract: In the process of gaining digital images whether scanning or photographing, the rugged manuscript or the nonlinear characteristic of the camera lens, etc. may lead to severe geometric distortions of the captured images. Therefore, geometric rectification of the distorted images is necessary before they can be further processed. In this paper, A careful analysis was done on the cases of geometric distortion, and then an auto-correction method is deduced. In the correction process for image scanned, global thresholding was used to extract the feature region which was needed, use Sobel operator to detect edge, then use Hough transform to auto-extract the boundary. After Hough transform, some parameters associated with the control point were obtained for correcting images. Extensive experiments proved that it can recognize the outline fast and correct geometric distortions effectively. The results show that the method was applicable and had high accuracy.

Abstract: Obstacle identification is one of the critical technologies of unmanned vehicle, edge detection is the basic step of obstacle identification based on video sensor and the magnitude guarantee of identification effect. In order to meet the demand of accuracy, real-time and stability of obstacle identification, a new multiple order morphology edge detection algorithm is proposed. We adopt two-dimensional histogram oblique segmentation to locate edge, then detected edge by improving the existing mathematical morphology edge detection operators and using appropriate structuring elements and percentile. Experimental results showed the edge detected is exquisite, continuous and intact. The algorithm possesses good robustness for different noised images, cuts operation time by nearly half compared with algorithm without edge location, then makes a good foundation for subsequent processing of obstacle identification.

Abstract: With the continuous development of video detection technology, the video analysis technology based on campus security has become an important part of the construction of safe campus. As the college students still are a group that has poor ability of security protection, campus security issue is closely related to the stability of society and family happiness, and has become a topic of concern to the whole society. The intelligent vision-based campus public safety monitoring system is an important means to achieve security monitoring, it can automatically analyze the video image sequence, and detect, track and identify objects in the monitoring scene without human intervention, and make high-level understanding and analysis of behaviors on this basis. Most of the existing visual monitoring systems can collect and store video data, and the real-time event detection task can automatically be generated through background analysis. Intelligent visual monitoring system should not only be used for accident investigation, but also be used to prevent potential disasters and accidents. The system is consisted of system management platform, event mining and analysis, monitoring and extraction of moving targets, forecasting and tracking targets. The paper makes an in-depth study on the application of intelligent visual detection technology on campus. Based on the intelligent visual video analysis, hidden Markov model is adopted in the paper for video event detection and analysis, motion features and shape features are taken as the observation data, and segmentation method is adopted to analyze the influence of video viewing height and angle on the detection result.