Research On Intelligent Wheelchair Control System And Image Recognition Technology

Along with the coming of aging population in China,the number of the elderly and disabled people is increasing.Wheelchair,as a kind of assistant tool for the disabled,will help the disabled to take care of themselves to a great extent,and help them to improve their freedom of action and reintegrate them into society.Using the technical innovation to create intelligent assisted wheelchairs suitable for the disabled and to meet their needs in production and life as far as possible will be an important research direction to help them.Aiming at the problem that disabled people can not operate wheelchairs for a long time,a kind of intelligent wheelchair control system based on image processing and multi-sensor fusion is designed in this paper.Compared with the ordinary wheelchair,the wheelchair added a camera module and WIFI module,the information can be fed back to the wheelchair control system in real time.the wheelchair can identify independent crosswalk on both sides of the road signs,pedestrian crossings and traffic lights,automatic driving.At the same time,the wheelchair can avoid obstacles in the course of driving,it can effectively solves the problem of inconvenient travel for the elderly and the disabled people.While the wheelchair driving on the road,the surrounding environment information is constantly changing,so under the influence of complex natural conditions,how to quickly and accurately identify the pedestrian crossing sign,pedestrian zebra crossing,traffic lights,is the focus and difficulty of this study.In order to overcome the influence of adverse conditions,this paper mainly does the following research:(1)The hardware design of wheelchair control system.According to the driving characteristics of the wheelchair,the hardware structure of the wheelchair is determined.Combined with the requirement of real-time image processing in wheelchair control system,the camera module,wireless transmission module and electronic compass module are added to the wheelchair to achieve real-time information transmission and wheelchair orientation display in wheelchair system,at the same time,it can improve the wheelchair's obstacle avoidance function.According to the actual needs of users and the characteristics of the hardware part of the control system,this paper further improves the software design of the control system,and introduces the function area of the interface of the software system.(2)Analyzing and processing the collected images.First,histogram equalization is used to improve the quality of the image.Through the experiments,several filtering algorithms are analyzed and compared,and the median filtering algorithm is used to eliminate the salt and pepper noise in the image.(3)To deal with the environmental variability of traffic signs,and at the same time,the real-time and accuracy of the image algorithm is improved.In this paper,color space conversion,threshold segmentation,morphological analysis,connected region and other algorithms are used to segment the image.Then the characteristics of the pedestrian crossing sign,pedestrians zebra line and traffic signal light area are extracted.(4)Aiming at the impact of traffic signs such as damage,slanting and uneven illumination,this paper applies pyramidal multi scale shape matching and traffic signs' own characteristics to design an image processing algorithm for recognizing crosswalk signs,pedestrian zebra crossing and traffic lights.The control system mainly uses the image processing software Halcon10.0 and C# language to complete the design of the image processing algorithm.The validity and feasibility of the algorithm are verified by the experimental data,and the image recognition of pedestrian crossing sign,pedestrian zebra crossing and traffic signal light is realized.