The Compact Spinning Lattice Apron Quality Inspection Based On Digital Image Processing

Lattice apron, an important component of compact spinning system, is widely used in the world. Because of its key effect on the yarn forming, its development and progress affect the developing trend of negative pressure airflow compact spinning technology. Nowadays, there is no appropriate standard system for the performance inspection of lattice apron, especially for the evenness of mesh distribution. Traditionally, the lattice apron quality inspection is evaluated visually, which is subjective and poorly precise. Computerized image processing techniques are now widely used in the inspection and quality control of textile industry, which can carry out transformation on the image and extract characteristic information from the image conveniently. Therefore it is a trend to use image processing techniques to evaluate the appearance of lattice aprons objectively.In this thesis, the objective assessment system of lattice apron quality inspection is based on digital image processing. The results of the inspection based on digital image processing are efficient and objective and will not be influenced by the environment. Image acquisition device is designed based on the feature of the lattice apron. CCD camera is used to acquire the image of two types of lattice aprons with warps and wefts in the same color or different colors. According to their features, different algorithms are used to preprocessing the images to get binarization images. The image of the lattice apron is segmented according to the midline of the warps and wefts. Each segmented mesh images is marked, and characteristic information of the sizes and shapes of the mesh images is extracted and stored. Finally, the quality of the lattice apron can be evaluated through the overall analysis of all characteristic values. The main conclusions of this thesis are briefly summarized as follows:(1) Based on digital image processing method, the assessment system of lattice apron quality has been developed. Which integrates the image processing and analyzing program with image-capturing device. The work principle of the image acquisition device, light sources and imaging principle has been analyzed to achieve the design and implementation of hardware. (2) Subtracting background technique is applied to eliminate the non-uniform illumination on images. According to the feature of different acquisition images, Linear transformation and histogram equalization is used to implement image enhancement. High frequency noise is eliminated by Wiener filter to accomplish image preprocessing. The binarization of grey-scale image of lattice aprons with warps and wefts of the same color is implemented by double-peak threshold value method. The image of lattice aprons with warps and wefts of different colors are filtered horizontally and vertically then integrated to get binarization image.(3) The binarization image is processed using smoothing filtering to get smoothing image on both warpwise and fillingwise. Then the image is segmented according to the center line of the smoothing image.The characteristic size and the shape feature (area, perimeter, length-width ratio of its minimum enclosing rectangle, rectangular degree, etc.) of each segmented mesh is extracted; The air voids and the total void count can also be calculated by general eigenvalue. The actual void amount in the testing image is 294, the same as the measuring value. The error between theoretical air voids 26.89%and the measuring value 23.90% is little, which means the eigenvalues acquired though image processing is feasible.(4) After the research on static algorithm, the Gui function of Matlab is used to write the static algorithm into software and realize the interactive inspection system.