Current Issue : October - December Volume : 2015 Issue Number : 4 Articles : 4 Articles
In this study, we present an application of neural network and image processing\ntechniques for detecting the defects of an internal micro-spray nozzle. The defect regions\nwere segmented by Canny edge detection, a randomized algorithm for detecting circles and\na circle inspection (CI) algorithm. The gray level co-occurrence matrix (GLCM) was\nfurther used to evaluate the texture features of the segmented region. These texture features\n(contrast, entropy, energy), color features (mean and variance of gray level) and geometric\nfeatures (distance variance, mean diameter and diameter ratio) were used in the\nclassification procedures. A back-propagation neural network classifier was employed to\ndetect the defects of micro-spray nozzles. The methodology presented herein effectively\nworks for detecting micro-spray nozzle defects to an accuracy of 90.71%....
This study proposes an automatic reading approach for a pointer gauge based on computer vision. Moreover, the study aims to\nhighlight the defects of the current automatic-recognition method of the pointer gauge and introduces amethod that uses a coarseto-\nfine scheme and has superior performance in the accuracy and stability of its reading identification. First, it uses the region\ngrowing method to locate the dial region and its center. Second, it uses an improved central projection method to determine the\ncircular scale region under the polar coordinate system and detect the scale marks. Then, the border detection is implemented\nin the dial image, and the Hough transform method is used to obtain the pointer direction by means of pointer contour fitting.\nFinally, the reading of the gauge is obtained by comparing the location of the pointer with the scale marks. The experimental results\ndemonstrate the effectiveness of the proposed approach. This approach is applicable for reading gauges whose scale marks are either\nevenly or unevenly distributed....
Development of FPGA-based, network-enabled embedded systems in Register\nTransfer Level hardware description languages is tedious. Despite the automation of this\nprocess with numerous EDA tools available, no well-established design patterns exist. Moreover,\nthe entire production cycle requires appropriate theoretical background and hardware\ndesign intuition from the developer which discourages the software community. To improve\nproductivity and minimize time to market when assembling such systems, we propose a new\nhardware/software co-design approach to building reconfigurable hardware web services.\nThe proposed integrated development platform features a programmable FPGA board where\ncomputations of different nature and purpose are logically distributed among a sequential\nsoft-core processor program, a massively parallel accelerator and an independent communication\nmodule that handles remote clients� requests. Our second contribution is a set of\ntools that make the development of the aforementioned services essentially a software design\nundertaking with the extensive use of high-level programming languages. The platform\nhas been tuned to act as a flexible runtime environment for image processing services, thus\nproviding functionality of an intelligent camera. Two example services built from scratch\naccording to the new methodology are discussed. Reduced development time and significant\nperformance gain observed prove validity of the proposed approach and unveil a large\npotential of the assembled prototype....
Based on the traditional machine vision recognition technology and traditional artificial neural networks about body movement\ntrajectory, this paper finds out the shortcomings of the traditional recognition technology. By combining the invariant moments of\nthe three-dimensional motion history image (computed as the eigenvector of body movements) and the extreme learning machine\n(constructed as the classification artificial neural network of body movements), the paper applies the method to the machine vision\nof the body movement trajectory. In detail, the paper gives a detailed introduction about the algorithm and realization scheme of the\nbody movement trajectory recognition based on the three-dimensional motion history image and the extreme learning machine.\nFinally, by comparing with the results of the recognition experiments, it attempts to verify that the method of body movement\ntrajectory recognition technology based on the three-dimensional motion history image and extreme learning machine has a more\naccurate recognition rate and better robustness....
Loading....