【摘要】：帶鋼產(chǎn)品,作為鋼鐵產(chǎn)品中的一個(gè)重要部分,已經(jīng)成為航空航天、機(jī)械制造、汽車(chē)生產(chǎn)、化工等工業(yè)的重要原材料,其質(zhì)量直接影響著產(chǎn)品的最終性能。因此,作為帶鋼表面質(zhì)量自動(dòng)評(píng)估的一種重要手段,基于機(jī)器視覺(jué)的帶鋼表面缺陷在線(xiàn)檢測(cè)的研究具有重要的理論與現(xiàn)實(shí)意義。針對(duì)現(xiàn)有常規(guī)缺陷檢測(cè)方法在實(shí)際應(yīng)用中吞吐量低、檢測(cè)準(zhǔn)確率不高的問(wèn)題,本文從基于紋理異常檢測(cè)和機(jī)器學(xué)習(xí)兩方面著手,分別提出了基于局部二值模式(LBP)的缺陷紋理異常檢測(cè)和基于卷積神經(jīng)網(wǎng)絡(luò)CNN的缺陷檢測(cè)方法。同時(shí),為了提高分類(lèi)算法的準(zhǔn)確性,提出一種基于特征對(duì)的改進(jìn)ReliefF特征選擇方法。本文研究的內(nèi)容與成果如下:(1)針對(duì)傳統(tǒng)檢測(cè)方法在帶鋼缺陷檢測(cè)中誤檢率和漏檢率較高,檢測(cè)算法參數(shù)較多的問(wèn)題,提出了一種基于多尺度LBP編碼的缺陷檢測(cè)方法。在不同尺度下對(duì)帶鋼表面圖像創(chuàng)建高斯差分金字塔模型,確定缺陷疑似區(qū)域,然后針對(duì)可疑區(qū)域進(jìn)行經(jīng)過(guò)閾值化處理后的LBP編碼,最后將所有尺度下的編碼圖像融合,生成最終的像素級(jí)缺陷位置信息,將檢測(cè)結(jié)果中的連通域合并(ROI合并),生成完整的缺陷位置信息。(2)為了排除背景紋理的干擾,抑制偽缺陷的產(chǎn)生,本文再次提出一種基于奇異值分解的缺陷檢測(cè)方法SVD-LBPH,通過(guò)對(duì)待檢測(cè)圖像進(jìn)行SVD分解與重構(gòu),弱化背景紋理,然后采用LBP對(duì)圖像進(jìn)行編碼,提取LBP直方圖相關(guān)的統(tǒng)計(jì)特征,通過(guò)將特征與設(shè)定的閾值比較,最終檢測(cè)出缺陷。(3)實(shí)時(shí)檢測(cè)對(duì)實(shí)時(shí)性要求較高,實(shí)時(shí)檢測(cè)結(jié)束后,需要在檢測(cè)出來(lái)的缺陷區(qū)域進(jìn)行即時(shí)檢測(cè),進(jìn)一步剔除偽缺陷。在即時(shí)檢測(cè)階段,針對(duì)深度學(xué)習(xí)算法在目標(biāo)檢測(cè)與圖像分類(lèi)的高準(zhǔn)確率,采用卷積神經(jīng)網(wǎng)絡(luò)CNN進(jìn)行缺陷的檢測(cè),通過(guò)與其它基于機(jī)器學(xué)習(xí)的缺陷檢測(cè)方法進(jìn)行對(duì)比,驗(yàn)證了卷積神經(jīng)網(wǎng)絡(luò)在缺陷檢測(cè)的高效潛力。(4)為了對(duì)缺陷類(lèi)別進(jìn)行辨別,本文提取了缺陷的灰度特征、灰度共生陣以及頻域等特征。為了提高分類(lèi)準(zhǔn)確率,剔除不相關(guān)的特征,同時(shí)為了避免因特征維數(shù)過(guò)大而造成的過(guò)擬合,采用了特征篩選的手段。本文利用更新特征對(duì)權(quán)重的方式對(duì)ReliefF進(jìn)行改進(jìn),實(shí)現(xiàn)特征的降維。最后利用SVM對(duì)篩選后的特征進(jìn)行分類(lèi)。
[Abstract]:Strip products, as an important part of iron and steel products, have become an important raw material in aerospace, mechanical manufacturing, automobile production, chemical industry, etc. The quality of strip products directly affects the final performance of the products. Therefore, as an important means of automatic evaluation of strip surface quality, the research of on-line detection of strip surface defects based on machine vision has important theoretical and practical significance. Aiming at the problems of low throughput and low detection accuracy of the existing conventional defect detection methods in practical applications, this paper starts from two aspects: texture-based anomaly detection and machine learning. Defect texture anomaly detection based on local binary mode (LBP) and defect detection based on convolution neural network CNN are proposed respectively. At the same time, in order to improve the accuracy of the classification algorithm, an improved ReliefF feature selection method based on feature pairs is proposed. The contents and achievements of this paper are as follows: (1) aiming at the problems of high false detection rate and high miss detection rate and more parameters of detection algorithm, a new defect detection method based on multi-scale LBP coding is proposed. Gao Si differential pyramid model is created for the strip surface image at different scales to determine the suspected defect area, and then the LBP coding after threshold processing is carried out for the suspected area. Finally, the coding image at all scales is fused. The final pixel level defect location information is generated, and the connected domain (ROI merging) in the detection result is combined to generate the complete defect location information. (2) in order to eliminate the interference of background texture, the false defect is suppressed. In this paper, a new defect detection method based on singular value decomposition (SVD) is proposed, which weakens background texture by SVD decomposition and reconstruction of detected image, and then uses LBP to encode the image and extract the statistical features related to LBP histogram. By comparing the features with the set threshold, the defects are finally detected. (3) Real-time detection requires high real-time performance. After the real-time detection, it is necessary to detect the defects in the detected areas immediately, and further eliminate the false defects. In the phase of immediate detection, aiming at the high accuracy of depth learning algorithm in target detection and image classification, a convolutional neural network (CNN) is used to detect defects, which is compared with other defect detection methods based on machine learning. The high efficiency potential of convolution neural network in defect detection is verified. (4) in order to distinguish the defect category, the gray level feature, gray level co-occurrence matrix and frequency domain feature of defect are extracted in this paper. In order to improve classification accuracy, eliminate irrelevant features, and avoid over-fitting caused by large feature dimension, feature selection is adopted. In this paper, ReliefF is improved by updating the weight of feature pair to reduce the dimension of feature. Finally, SVM was used to classify the selected features.
【學(xué)位授予單位】：河北工業(yè)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2015
【分類(lèi)號(hào)】：TG142.15;TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 徐科;宋暢;;基于全局二值模式的特征提取方法及其應(yīng)用[J];模式識(shí)別與人工智能;2013年09期

2 陳躍;張曉光;;帶鋼表面缺陷圖像的可拓分類(lèi)算法[J];計(jì)算機(jī)工程與應(yīng)用;2013年21期

3 宋克臣;顏云輝;彭怡書(shū);董德威;;引入局部信息的帶鋼缺陷圖像凸優(yōu)化活動(dòng)輪廓分割模型[J];機(jī)械工程學(xué)報(bào);2012年20期

4 許豪;孔建益;湯勃;王興東;劉源l，

本文編號(hào)：2385177