充分發(fā)揮圖像傳感器在工業(yè)自動(dòng)化中的作用

　　成像正在成為許多嵌入式系統(tǒng)發(fā)展的關(guān)鍵部分。本文著眼于圖像傳感技術(shù)——從傳感器到算法和IP——越來(lái)越多地用于工業(yè)視覺(jué)。

　　圖像傳感器在工業(yè)自動(dòng)化領(lǐng)域的應(yīng)用越來(lái)越廣泛。成熟的CMOS技術(shù)提供了更大的傳感器陣列以提供更大的視場(chǎng)，以及更快的圖像捕獲以滿(mǎn)足視覺(jué)檢查的要求。與此同時(shí)，新的成像架構(gòu)和新的軟件算法提高了識(shí)別速度，并使系統(tǒng)更具成本效益。隨著現(xiàn)場(chǎng)可編程門(mén)陣列（FPGA）密度的增加和成本的降低，這些算法在工廠中的應(yīng)用越來(lái)越廣泛，同時(shí)也幫助了圖像傳感器的普及。傳統(tǒng)的觀點(diǎn)認(rèn)為機(jī)器視覺(jué)需要最高的分辨率（見(jiàn)表1），CMOS器件的新架構(gòu)提供了更大的性能，而標(biāo)準(zhǔn)CMOS器件被用于整個(gè)工廠的更多成像和自動(dòng)化應(yīng)用。

　　CMOS sensors are the mainstay of industrial imaging. Although they are not as high resolution as bipolar charge coupled devices （CCDs）， they are cost effective and rugged for use in industrial imaging. The same ruggedness and reliability that is being applied to designs in the automotive market for applications such as lane detection is now becoming available for industrial vision. While there has been a performance advantage for CCD in industrial vision， the reducing cost of mainstream sensors and cameras is allowing vision to be used cost effectively on the factory floor， and the increasing performance of high-end CMOS sensors is competing directly with CCD technology.