摘要
目前循迹智能车的跑道识别都依赖于跑道上的特殊标志,一旦识别失败,小车将脱离原有道路。针对该现状,提出一种基于图像特征提取的循迹方案。该方案将循迹信息与跑道分离,循迹系统由信息采集和动作执行两部分组成。将目标路线图画在透明压克力板上,通过CCD摄像头采集图像,利用MC9S12XS128单片机对图像数组进行边缘提取和角点检测得到路径信息,然后将路径信息经无线模块NRF24L01发送给智能车,通过精确的缩放比例将绘图区域映射到实际路径,完成分离循迹。行驶过程中无需判别跑道特征。
Nowadays, tracking intelligent car＇s runway identification depends on tile special sign on it. Once the identification fails, the car will be uncontrollable. Aiming at this situation, a method based on the feature extraction is proposed in this paper . The scheme separates the tracking information from runway, which consists of information collection part and execution part. The route chart is painted on a transparent acrylic board, which will be captured by the CCD camera. Then the MCU MC9S12XS128 will recognize the routing information by edge extraction and corner detection. After that, routing information is sent to the intelli gent car by wireless module NRF24L01. To complete the separation-tracking, the drawing area is mapped to the actual route through the precise scaling. There is no necessary to distinguish the feature of runway during the moving process.
出处
《微型机与应用》
2013年第10期84-86,共3页
Microcomputer & Its Applications
基金
国家自然科学基金(61271334)
