Design And Implementation Of Information Collection And AI Interaction System Based On Raspberry PI

With the quickening of the process of information and intelligence in social life,people's demand for exploring information is becoming more and more intense.Information is provided and collected everywhere in our life,especially after the outbreak of new coronavirus,information has become a kind of “Pass” to some extent,at present,to a large extent,it is still the collection and processing of information manually.In addition,some important information in complex and harsh environment such as industry and field is not suitable to be collected manually,so this paper designs an interactive system of information collection and AI based on raspberry,with the support of OPENCV image database,the system uses the combination of software and hardware to recognize and process the effective information of objects,faces and colors in the collected images,and provides feedback through speech synthesis and broadcast,the aim is to complete the whole process of information collection,processing and feedback intelligently and efficiently.Starting from the practical significance of information collection and processing,this paper briefly introduces the related technologies involved in the system,expounds the research significance of the information collection and Ai Interactive System,and then introduces the theoretical basis of the system;The design and implementation of the system is divided into two parts: one is the hardware part,the system takes Raspberry Pie as the core processor,and combines the Omnidirectional mecanum wheel and the 1,2 degree of freedom high definition camera which can manually lift and lower the head,as a mobile collecting terminal,the system can collect information flexibly.Secondly,in the software part,the system takes Jupyterlab as the development tool,uses python 3 as the main programming language,and preprocesses the collected image information by calling the library function in OPENCV,haar cascade classifier,Tensor Flow training object detection model and HSV color space are used to recognize face,object,color and so on.In addition,the system uses PID to adjust the motor,effectively controls its movement and realizes the camera to track the target in the largest range,realizes the real-time detection to the target in the picture,and then uses the TTS technology to synthesize and broadcast the information.After completing the design of the system,this paper tests and analyzes each function of the information acquisition and Ai Interactive system repeatedly,and sums up the shortcomings of the system as a whole and the merits of improvement,the future upgrading of the system is prospected.The final test results can show that the system is convenient,efficient and adaptable,and can display and process the environmental information accurately.It is worth mentioning that the system also shows great advantages in omni-directional wheel and voice broadcast,which makes the system more efficient,convenient and practical.