李白燕 張健
摘要:為了提高出廠產(chǎn)品的質(zhì)量,需要在生產(chǎn)線上對產(chǎn)品進行高速在線檢測,該文采用基于機器視覺的檢測系統(tǒng)來進行在線檢測產(chǎn)品疵點。該系統(tǒng)采用類別共生矩陣算法建立紋理模型,采用CCD視覺傳感、高速DSP芯片、USB數(shù)據(jù)傳送等技術結(jié)合軟件算法構成高速在線機器視覺檢測系統(tǒng)。
關鍵詞:在線檢測;產(chǎn)品疵點;數(shù)字信號處理器;線陣CCD;USB控制器;類別共生矩陣
中圖分類號:TP391文獻標識碼:A文章編號:1009-3044(2012)15-3572-02
Based on DSP Online Detection System for Products Defect
LI Bai-yan, ZHANG Jian
(School of Information Engineering, Huanghuai University, Zhumadian 463000, China)
Abstract: In order to improve the quality of the products before leaving the factory,need high speed online inspection for the products in the production line, the paper uses detection system for online detection products defect based on machine vision. The label co-occurrence matrix algorithm establish texture model, using CCD(Charge Coupled Device), high speed DSP chip, USB data transfer technology com bined a with software algorithmic gives the realization method of high-speed on-line machine vision measuring system.
Key words: online detection ; products defect; DSP ;linear CCD; USB; label co-occurrence matrix
隨著計算機技術、電子通信技術、數(shù)字信號處理技術等硬件的發(fā)展,軟件算法的不斷完善,計算機視覺技術日臻成熟,使自動化檢測技術趨向高精度、高效率、高可靠性,日益代替人工檢測方法,大大提高了出廠產(chǎn)品的質(zhì)量,降低了生產(chǎn)成本,提高了生產(chǎn)效率。本文根據(jù)產(chǎn)品紋理圖像的特點,采用類別共生矩陣算法(FLCM)對產(chǎn)品疵點進行產(chǎn)品疵點檢測,該算法具有計算量小,能突出細微疵點等優(yōu)點。采用高速電荷耦合器件CCD TCD1206、高速DSP5409、USB控制器AN2131Q及軟件算法硬件實現(xiàn)技術等,建成在線檢測產(chǎn)品疵點系統(tǒng)。本文所建立的系統(tǒng)不僅在檢測的準確率上有很大的提高,而且還大大地提高了檢測的效率,減少了成本,這些都促進了在線自動實時檢測的應用。