Applied Mechanics and Materials Vols. 536-537

Abstract: In a condom sorting process, we need to shoot the state image of condom through the camera and do the ellipse detection to determine the front position of condom on the conveyor belt, so developing the image processing algorithms which can detect the condom ring is necessary. The algorithm based on the Hough circle detection can obtain better detection effect for the most forms of the condom ring position, but it can not correctly detect the circle when the image shows linear shape; the algorithm based on the contrast of the ring detection get better effect on the ring in the image showing linear, while it has a low recognition rate in more folded case. We combined the two algorithms to use, proposed a combined algorithm which overcame the respective shortcomings of the two algorithms. Verified by experiment, the results of detection, recognition rate and the average processing time show the combined algorithms get the best comprehensive effect.

Abstract: Taking the central heating of Caofeidian for example, this paper introduces the video transmission system which bases on 3G wireless networks and the realization of communication. The video images are collected in remote heat-exchange station and are transmitted to the monitoring and controlling center through 3G wireless network. The result of operation indicates that the system which meets the requirements of long-term stable operation of remote monitoring is not only easily operated, but also with high regulation accuracy and stable.

Abstract: In view of the authentication model based on token and password of traditional web cloud storage easily is forgotten or theft that raise questions about identity theft. This paper proposes a identity authentication scheme of web cloud storage based on fingerprint identification. Using the identity authentication model of fingerprint identification to verify web cloud storage users that satisfy people to the web of cloud storage security, reliability and convenience of the proposed requirements, effectively solve the problem of the resource of unauthorized access. The paper put forward three layers of c/s architecture that effectively improve the flexibility, scalability and efficiency of authentication system. And mixed encryption algorithm is put forward as two-way authentication identity authentication protocol improve system security and realize high transmission of huge amounts of data. Finally through the test and analysis of practical application proves that the security authentication based on fingerprint identification applying in the web cloud storage certification system is feasible.

Abstract: The equipment voice de-noising and reconstruction algorithm is presented in this paper based on intelligent robot in transformer situation (transformer and high resistance). The method is an embodiment of information and intelligence which can be applied in voice recognition of inspection robot. Firstly, equipment voice was collected and formed a sample database. Secondly, the voice features were analyzed in sample database, and the power-frequency Fourier spectrum as the voice features was extracted in the frequency domain. Finally, Fourier spectrum of power-frequency multiple was used to construct the voice by taking advantage of the inverse Fourier transform. The experimental results show that the algorithm has stable performance, strong robustness, and can accurately reconstruct the voice, so it is benefit to the follow-up voice recognition.

Abstract: In this study, a novel two-phase image segmentation algorithm (TPIS) by using nonlocal mean filter and kernel evolutionary clustering in local learning is proposed. Currently, the difficulties for image segmentation lie in its vast pixels with overlapping characteristic and the noise in the different process of imaging. Here, we want to use nonlocal mean filter to remove different types of noise in the image, and then, two kernel clustering indices are designed in evolutionary optimization. Besides, the local learning strategy is designed using local coefficient of variation of each local pixels or image patch is employed to update the quality of the local segments. The new algorithm is used to solve different image segmentation tasks. The experimental results show that TPIS is competent for segmenting majority of the test images with high quality.

Abstract: Object association and tracking have attracted great attention in the computer vision. In this paper, we present an object association and tracking method for monitoring multiple vehicles on the road based on objects' visual features and the similarity comparison between them. First, we identify vehicles using the difference operation between the current frame in CCTV image sequences and the referential images that are stored in a database, and then extract various features from the vehicles identified. Finally, we associate the objects in the current frame with those in the next frames using similarity comparison, and track multiple objects over a sequence of CCTV image frames. Empirical study using CCTV images shows that our method has achieved the considerable effectiveness in tracking vehicles on the road.

Abstract: In the computer vision, feature extraction is an important research orientation. Corner and edge detection is a basic technique for obtaining local feature in this image. In the paper, we propose a method that corner and edge of image is detected simultaneously by an annealed chaotic competitive network. The method is compared with several traditional methods of corner and edge detection algorithm, the experiment result shows a good performance. The clustering method can achieve the corner and edge detection simultaneously.

Abstract: This paper describes the image based on dark channel prior to fog. Dark colors from the statistical laws of outdoor priori no fog image database, it was such a key observation was based on the fact that - the vast majority of each local area outdoor image without fog are present strength of certain of at least one color channel low value pixels. Using this model building, directly estimate the concentration of fog and mist removal recover high quality interference image. Experiments show that the dark channel prior to remove the image haze becomes simple and effective.

Abstract: Optical information processing systems have been widely applied, this paper studies the calculation of the amount of a reduction in the use of genetic algorithms to increase the diversity of the sample algorithm to reduce problems caused by the sample dilution with a resampling method, combined with the finite field resampling, resampling constitute a finite field of genetic algorithms. We using surface to capture the dynamics of template update, to prevent inappropriate tracking results into the template set, pending the outcome of the tracking template is added to the first test for cover. Detect occlusion control template updates. Prevent tracking results of drift.

Abstract: This paper focuses on the visual tracking algorithm in optical imaging surveillance and tracking system. The tracking particle filter framework deemed find sparse representation problem, can effectively overcome the visual tracking algorithm appears in noise, occlusion, background interference and complex situations such as illumination changes. Morphological methods using digital occlusion area is detected to determine whether the date is added to the template tracking results set, thereby updating the control template, to effectively prevent the drift tracking results.