一种图像识别车辆定位系统摄像头的设计  

Design of a Camera for Image Recognition Vehicle Positioning System

作　　者：宋昱晓 王雨璐 于宏博 王永恒 王宪彬 Song Yuxiao;Wang Yulu;Yu Hongbo;Wang Yongheng(Northeast Forestry University,Harbin 150000,China)

机构地区：[1]东北林业大学,哈尔滨150000 [2]不详

出　　处：《黑龙江科学》2023年第4期154-155,158,共3页Heilongjiang Science

摘　　要：为满足人们的停车需求,解决大型停车场停车难与找车难问题,提出了一种图像识别车辆定位查询的新型摄像头,它能够对摄像头主体前端的镜头进行清洁,避免尘土及其他杂质附着在镜头表面而降低摄像头主体图片采集的清晰度,通过训练此摄像头建立本地模型,结合百度AI云平台构造云端模型,协助车主在短时间内定位车辆,规划路线,实现反向寻车,加快停车场内外周转,有利于提高停车用户的满意度,促进停车场的智能化管理。In order to better meet the parking demand,and solve the problem of the difficulty in parking and finding vehicles in large-scale parks,the study proposes a new type of camera for vehicle location query based on image recognition,which not only cleans the lens at the front of the camera body to prevent dust and other impurities from adhering to the lens surface and reducing the sharpness of the camera body image collection,but also helps owners to locate their vehicles in a short time by training this camera to build a local model through a cloud model of combining with Baidu AI cloud platform.The planned route realizes reverse searching,speeds up the turnover inside and outside the parking lot,improves the satisfaction of parking users,and promotes intelligent management of the parking lot.

关 键 词：大型停车场 图像识别 清洁除尘 百度AI 摄像头 

分 类 号：TN925[电子电信—通信与信息系统] U491.71[电子电信—信息与通信工程]