基于视觉显著性的太阳能电池片表面缺陷检测

发布时间：2018-04-20 05:24

本文选题：太阳能电池 + 表面缺陷检测　；参考：《仪器仪表学报》2017年07期

【摘要】：现有基于机器视觉的太阳能电池片表面缺陷检测算法均是采用各种类型的数学模型来进行算法设计,为进一步提高检测准确率,从人眼仿生学角度出发,首次将人眼的视觉注意机制引入到太阳能电池片表面缺陷检测中,提出了一种基于视觉显著性的太阳能电池片表面缺陷检测算法。首先,对输入的太阳能电池片表面图像进行预处理,去除对检测有影响的噪声和栅线;其次,提出一种基于自学习特征的视觉显著性检测算法来大致定位缺陷区域;随后,提出一种视觉显著性和超像素分割相结合的算法来进一步精确定位缺陷区域;最后,通过形态学后处理得到最终检测结果。在包含多种缺陷类型的测试图像库上的主观和客观实验评估表明,该算法具有较高的检测准确率。
[Abstract]:The existing solar cell surface defect detection algorithms based on machine vision are designed with various mathematical models. In order to further improve the accuracy of detection, from the point of view of human eye bionics, The visual attention mechanism of human eyes is introduced into the surface defect detection of solar cells for the first time, and an algorithm based on visual saliency is proposed to detect the surface defects of solar cells. Firstly, the input solar cell surface image is preprocessed to remove the noise and grid lines that affect the detection. Secondly, a visual saliency detection algorithm based on self-learning feature is proposed to roughly locate the defect area. A new algorithm combining visual saliency and hyperpixel segmentation is proposed to locate the defect area accurately. Finally, the final detection results are obtained by morphological post-processing. The subjective and objective experiments on the test image database with various defect types show that the proposed algorithm has a high detection accuracy.
【作者单位】：郑州轻工业学院电气信息工程学院;
【基金】：国家自然科学基金(61501407,61503173) 河南省高等学校重点科研项目(15A413006) 河南省重大科技专项(161100211600) 河南省科技厅科技攻关项目(162102210060,172102210062) 郑州轻工业学院博士基金(2014BSJJ016,2016BSJJ006,2016BSJJ002);郑州轻工业学院研究生科技创新基金(2016026)项目资助
【分类号】：TM914.4;TP391.41

【相似文献】