發(fā)布時(shí)間：2018-06-01 00:14

  本文選題：深度學(xué)習(xí) + 卷積自編碼器　； 參考：《電子測量與儀器學(xué)報(bào)》2017年08期

【摘要】：針對傳統(tǒng)織物缺陷檢測手工提取特征困難,疵點(diǎn)樣本有限的問題,結(jié)合卷積自編碼器(CAE),提出一種基于Fisher準(zhǔn)則的棧式去噪自編碼器算法(FSDAE)。首先從原始圖像中截取若干小塊圖像,采用稀疏自編碼器(SAE)訓(xùn)練,得到小塊圖像的稀疏性特征;其次利用該特征,初始化CAE網(wǎng)絡(luò)參數(shù),提取原始圖像的低維特征;最后將該特征數(shù)據(jù)送入FSDAE網(wǎng)絡(luò)進(jìn)行疵點(diǎn)檢測分類。分別對3類織物進(jìn)行測試,實(shí)驗(yàn)結(jié)果表明,算法能夠有效地提取織物圖像的分類特征,且通過加入Fisher準(zhǔn)則,提高了織物疵點(diǎn)的檢測率。
[Abstract]:Aiming at the problem of traditional fabric defect detection which is difficult to extract features manually and limited defect samples, a stack de-noising self-encoder algorithm based on Fisher criterion is proposed. Firstly, some small images are intercepted from the original image, and the sparse self-encoder is used to obtain the sparse feature of the small block image, and then the CAE network parameters are initialized to extract the low-dimensional feature of the original image. Finally, the feature data is sent into FSDAE network for defect detection and classification. The experimental results show that the algorithm can effectively extract the classification features of fabric images, and the detection rate of fabric defects is improved by adding Fisher criterion.
【作者單位】： 西安工程大學(xué)電子信息學(xué)院;
【基金】：國家自然科學(xué)基金(61301276) 陜西省工業(yè)科技攻關(guān)項(xiàng)目(2015GY034)資助
【分類號】：TP391.41;TS101.97

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 嚴(yán)平;鄧中民;劉童花;;基于改進(jìn)的小波分解織物疵點(diǎn)檢測[J];紡織科技進(jìn)展;2007年04期

2 餳谷`欠，

本文編號：1961983