Research On Image Recognition System Of Fully Automatic Light Inspection Machine Based On FPGA And DSP

Infusion drug products contain glass debris, hair, fibers, and other foreign matter in the manufacturing process. Infusion security issue has been a topic of common concern. Order to reduce the damage to the personal safety of medical infusion drugs, in the Infusion pharmaceutical production process, the national pharmacopoeia specified visible foreign matter may not be detected. But many of traditional medical infusion bottle detection method is manual inspection, artificial low detection efficiency, a narrow range of low accuracy, detection effect is unstable. The introduction of automatic light inspection machine, you can greatly improve the the infusion drug testing accuracy and efficiency, reduce labor costs, but also improve the drug safety index, reduced foreign impurities in the liquid drug to cause harm to the patient’s body.Automatic light inspection machine objective is to detect and identify the content of this thesis is based on FPGA and DSP automatic light inspection machine image recognition system, the system uses digital image processing technology to medical infusion drug inside the particles and other impuritiesinfusion drugs with a wide range of real-time detection and identification, and the failure of the product to the automatic light inspection machine removed the implementing agencies. The main contents are as follows:(1) Image recognition system architecture design and hardware implementationFirst describes the working principle of the automatic light inspection machine, fully automatic light inspection machine features and performance indicators to determine the system design of the image recognition system, and then introduced the hardware design of the image recognition system.(2) Image recognition system software designFirst introduced the system software development process and development environment. Second, according to the system functional requirements and FPGA, DSP work on the system software design process are described. The initialization of the image sensor, image acquisition and pre-processing, the system displays the function is completed by the FPGA; system initialization and identification algorithm implementation is done by the DSP.(3)System simulation and debug analysisFirst the various functional modules of FPGA and DSP system architecture for testing and simulation, to meet the performance requirements of the system experiments and commissioning of the entire system. To analyze the experimental results by specific experimental operation of the system effect, to identify problems and propose an improved method.This thesis is based on the design of the FPGA and DSP automatic light inspection machine image recognition system, after a large number of experimental data and debug simulation results verify that the system is designed to meet the requirements of automatic light inspection machine image recognition system.