FPGA-based Detection System Of Viscose Filament Yarn Blowing

Viscose filament because of its unique advantages in the textile industry has been more widely applied, but the quality of viscose filament instability has been affecting the development of an important factor, which is mainly caused by filament monofilament fabric dyeing caused by lack of uneven, now basically dependent on artificial completed viscose filament testing and quality control, the existence of this detection method is slow, inefficient and subject to human factors, low accuracy and can not reproduce and so many ills. In recent years, the emergence of machine vision and wide application of viscose filament yarn provided for the detection of a new idea, machine vision inspection system as a new detection technology, with higher speed, high efficiency, high precision, information obtained rich, low operating cost, simple and reproducible and so on. Effectively overcome the lack of the manual inspection, so as to achieve the automation of chemical fiber plant, production efficiency and product quality to provide a guarantee.Machine vision systems are characterized by increased production flexibility and automation. In some not suitable for manual operation of the hazardous work environment or artificial vision can't meet the requirement, the commonly used machine vision to replace the artificial vision; Tongshi in the process to industrial production, using artificial vision check product quality and low efficiency and high precision , machine vision inspection method can greatly improve production efficiency and production automation. And easy to implement information integration vision is to achieve the basis for computer integrated manufacturing technology.In this paper, blowing silk viscose filament detection system detection methods and related equipment and hardware design of a detailed discussion. Structure of the system finish in the second chapter, said. The system uses a blowing wire system is simple, with high accuracy to do so, and because the hardware is the use of FPGA-based embedded systems, so the overall system to achieve a dedicated and miniaturization. From now on test results can be seen, the proposed system meets the business requirements of the indicator, the system of error control within 2%. The main work of this paper is as follows:⑴introduced the economic value of viscose filament yarn, and put forward the purpose and significance of the detection. Further paper on the domestic development of viscose filament testing were discussed; the results of existing technology, proposed to replace the manual machine vision methods to detect, through lists of machine vision technology widely used test to prove that the use of machine vision viscose filament detection technology to solve the problem of feasibility.⑵in accordance with the requirements of the project proposed FPGA-based detection system viscose filament blowing the basic principles; and lists the hardware details of the various cell parameters indicators; rendering the system hardware structure, illustrates the hardware process operation.⑶details of the blow wire system design principles, including a gas box design and the silk board design, focuses on the aerodynamics of the system and, through a combination of experimental and theoretical find the most suitable system parameters.⑷on machine vision in optical imaging module of the design principles are introduced, which includes a light source of the design principle, the principle of microscope, camera lens of the imaging principle, CCD works. Further analysis of the reasons for the selection of the components and lists their parameters.⑸to achieve miniaturization of the system, first on the FPGA and the DSP works made us, and then applied to determine the parameters of the system chip. The system uses the combination of DSP and FPGA embedded system (later referred to as image processing FPGA board), because the FPGA logic processing in a great advantage, so that the image stored in the great advantages of speed, coupled with dedicated The image processing DSP (TMS320DM642) allows the system image processing (recognition and memory) has a fast and flexible software upgrade when the real advantages. Further from the detection system requirements, and discussed in detail the hardware design principles, and introduces a system of four modules: image acquisition module, image storage module, image analysis and processing module and Communication Module,And gives the detailed work of the circuit system.⑹the other hand, as the center of the image detection system processing section first discusses the software design, and detailed talks algorithm image processing system. Algorithm includes two aspects: image enhancement and counting. Viscose filament image enhancement methods include:①The principle of geometric transformation and application of pseudo-color enhancement image processing principles;②Introduced the commonly used method of image segmentation, and select the application of this system the threshold segmentation method ;③Introduced binary image processing basic geometric concepts, filament image information extraction methods to identify and distinguish the noise point of filament;④Introduced the morphology of the basic concepts and operations; for the conduct of a single filament tracking algorithm of image thinning algorithm was devoted.⑺Finally, viscose filament detection system based on the conclusions and recommendations put forward areas for improvement and related methods.