Textile Fabric Structure Identification System Research

Automation of textile production will be for a long period of time necessary to discuss the topic. Automation of textile production, including the first identification of the organizational structure of the fabric automation, automatic identification of the organizational structure of the fabric of textile production will play a significant role.The automatic identification of the organizational structure of the fabric with a full range of analysis and research, and finally in the Automatic Identification of fabric structure on the DM642image processing platform, practice has proved the DM642fabric image recognition, and effective, with widely The value of the program. The fabric structure identification including fabric image processing (early and late), the fabric image pre-processing includes fabric image tilt correction and stitching, and a major impact on the fabric image post-processing accuracy. Fabric image tilt by objective factors caused by the fabric of yarn tilt and image acquisition. To solve this problem, this paper presents a fast, high accuracy and the calculation of a relatively small amount of fabric image tilt correction algorithms, a better balance between the amount of computation and accuracy than previous methods. Fabric image field of view to be able to meet the cycle at least one organization, collected pieces of the same size of the overlap region of image stitching, this paper summarize the advantages and disadvantages of a variety of algorithms is presented based on Harris corner extraction fabric image stitching algorithm and improved the weighted average fusion algorithm for image stitching and crevices to eliminate, clear treatment effect, and the latter part of the fabric image processing. Image post-processing in the fabric, the fabric structure parameters to identify fabric images the autocorrelation function of detection organizational cycle; wavelet transform image decomposition fabric image vertically, horizontally details of the sub-image, to obtain the warp and weft density, the use of The parameters of the relationship between other parameters. The experimental results show that the original organization, changes in the organization has a high recognition rate. By this stage fabric point algorithm analysis and research, this article thoroughly discusses the advantages and technical difficulties of each algorithm, the integration of environmental considerations in the recognition of fabric weave point proposed a primitive matching identification improve the fabric image recognition method, a combination of algorithms and conventional algorithms effectively extends the algorithm to adapt range.Finally, computer development environment, the fabric a simple algorithm to identify the organization’s development, and then into a CCSC language development environmentrecognition algorithm, the results show that, with the predictable result has the same recognition effect, this is an embedded platform implemented on an attempt to identify the fabric image of great significance.