Research On The Pedestrian Detection Based On Radar And Video

At present,with the continuous improvement of computer performance and sensor technology,artificial intelligence and pattern recognition has been improved,all kinds of intelligent facilities are being fully entered into people's lives.They search for the direction and location of people,understand people's actions and language,and provide a wide variety of services.Pedestrian detection technology can help these intelligent facilities to analyze and judge whether there are people in the field of vision.Most of the traditional detection techniques rely on video images,and it is difficult to obtain 3D information from 2D image information,so it is unable to effectively deal with the longitudinal motion and non rigid motion of pedestrians in the scene.From the perspective of multi sensor fusion,this paper proposes radar as an additional information source of pedestrian detection,making up for the shortage of video images,and finally improving the accuracy of pedestrian detection.The main work of this paper includes the following two parts:In view of the specific needs of the airport,dam,frontier and other key sites,a new method based on video is proposed.In this method,the background environment of the scene is modeled by the Gauss mixture model.after the moving object is separated from the scene,the support vector machine is used to classify it.And the innovation is to make the pedestrian sample into standing posture and squatting two small class,in the training and classification process we regard the pedestrian sample as two different categories for processing,and finally unified output for pedestrians.In this way,the non rigid motion can be effectively improved,and the detection rate is improved.On the basis of the above,the method of using radar to improve the original method is put forward.We use radar to obtain the target distance,and then convert it into the actual area of the target,then use it as a feature for classification.In this way,the radar is used as a supplementary information source,and the description of the feature is enhanced,the basis of decision making is enriched,as well as effectively distinguish the shape of different objects,further improving the detection rate.Overall,the main research work of this paper is to focus on the research of the pedestrian detection technology which is integrated with the radar and video.It has important reference value and application value for the development of our country's development and related fields.