Design Of Monitoring Module Based On FPGA And Its Application On Intelligent Medicine Box

Recently,with the increasing number of chronic diseases patients such as tuberculosis,it is becoming more and more important to help patients with intelligent AIDS.Young people have the pressure of work,the busy life let them have no time to take medicine.The elderly as a special group,the body function and cognitive ability gradually weakened,which will cause their daily medication to cause obstacles.In the hospital,the nurse must go through every sickbed to remind the patient to take medicine,because the patient needs regular fixed-point and quantitative taking medicine.Their efficiency is low.Although many drug boxes have classified prompting functions now,few intelligent drug boxes have feedback function,that is,intuitively feedback images of patients taking medicine,so that doctors and family members can confirm that patients have taken medicine.Therefore,this article begins with the design and research of the intelligent drug box monitoring module.This thesis uses Verilog language to complete the basic function design,complete the basic operations such as shooting,screenshots,storage,display and so on,and lay a good hardware foundation for subsequent functions.Complete the OV7670 camera initialization and real-time shooting,and transmit the captured data to FPGA.The control circuit of the main memory SDRAM is completed to read,write and automatically refresh the data.Write asynchronous FIFO as the cache for data transmissions across time domain.Drive the VGA display to display and intercept the image function.This thesis studies a function that drives the camera to track the skin automatically.It is convenient for the inconvenient patients to complete the follow-up face recognition.Comparison of several detection algorithms based on skin models and experiments.Complete the processing of the color space conversion algorithm.Skin pixels in horizontal and vertical two directions are tracked with skin pixel location method,and the experimental verification is carried out.Considering that the method is limited in direction,we attempt to introduce fuzzy set theory into image processing and tracking direction determination,and propose a direct direction decision method with fuzzy sets.Experimental verification shows that it can accurately determine the location of skin area.When the position of the skin changes,the direction can be more smooth.In this thesis,the face recognition function is added to the intelligent drug box to prevent the patient from taking the wrong or forgetting the medicine.A variety of face recognition algorithms are studied.The image is preprocessed by the pixel gray algorithm,and the quality of the image is improved,and the face recognition function based on FPGA is realized.The experimental results show that the successful recognition rate is nearly 90%,and it is of practical value.Finally,this thesis analyzes the main basis and main principles of the selection of face box after face recognition,and studies the judgement method of opening the medicine box under various circumstances,and completes the selection logic and function realization of the medicine box.