电子器件
電子器件
전자기건
JOURNAL OF ELECTRON DEVICES
2014年
2期
315-321
,共7页
视觉导航%人工路标%交比不变性%PCA%最小距离分类器
視覺導航%人工路標%交比不變性%PCA%最小距離分類器
시각도항%인공로표%교비불변성%PCA%최소거리분류기
vision navigation%artificial landmark%invariance of cross-ratio%PCA%minimum distance classifier
针对以人工路标作为引导标识的视觉导航技术，提出了一种基于导引灯的人工路标识别算法。首先利用基于灰度信息的识别算法对路标上导引灯进行识别与定位；然后，利用交比不变性进行数字的初步定位；最后用投影法进行数字分割，主成分分析法( PCA)和最小距离分类器的模式识别方法对数字进行识别。算法在ICETEK-DM6437-B评估板上实现。实验表明，该算法具有很高的稳定性并能较好地满足实时性的要求。
針對以人工路標作為引導標識的視覺導航技術，提齣瞭一種基于導引燈的人工路標識彆算法。首先利用基于灰度信息的識彆算法對路標上導引燈進行識彆與定位；然後，利用交比不變性進行數字的初步定位；最後用投影法進行數字分割，主成分分析法( PCA)和最小距離分類器的模式識彆方法對數字進行識彆。算法在ICETEK-DM6437-B評估闆上實現。實驗錶明，該算法具有很高的穩定性併能較好地滿足實時性的要求。
침대이인공로표작위인도표식적시각도항기술，제출료일충기우도인등적인공로표식별산법。수선이용기우회도신식적식별산법대로표상도인등진행식별여정위；연후，이용교비불변성진행수자적초보정위；최후용투영법진행수자분할，주성분분석법( PCA)화최소거리분류기적모식식별방법대수자진행식별。산법재ICETEK-DM6437-B평고판상실현。실험표명，해산법구유흔고적은정성병능교호지만족실시성적요구。
In view of the vision navigation technology of using artificial landmark as the guiding identification,one recognition algorithm of artificial landmark based on the guiding light was proposed. Firstly,the algorithm based on gray information is used to recognize and locate the guiding light in the landmark. Then,the algorithm based on in-variance of cross-ratio realizes the localization to characters in the artificial landmark. At last,projection method is used to segment characters, meanwhile, both principal component analysis ( PCA ) method and minimum distance classifier are used to recognize the character. The algorithm is realized on the ICETEK-DM6437-B system board. The experiment shows the algorithm has high stability and meets the requirements of real-time well.
