摘要
在利用图像特征提取方法识别锯材表面缺陷的研究过程中,视觉系统中光源照射是评价图像质量的重要影响因素之一。为了得到更好的图像数据,采用主客观相结合的综合评价方法,通过设置光源工作距离为60,80和100 mm,分别从图像峰值信噪比、信噪比、对比度以及图像直方图4个角度探究图像质量的变化。选择红色和白色LED条灯,通过图像质量评价方法分析和实验验证相结合,进行视觉系统在不同光敏感程度、不同光照强度与不同光照频率下的成像效果差异性试验研究。试验结果表明,锯材活节的图像质量受光照强度的影响虽不显著,但光源在60 mm工作距离条件下得到的图像质量较好;由于锯材死节外沿偏黑的特殊性,当光源工作距离在高点与低点时,死节的图像质量变化不大,但应避免在80 mm条件下进行图像获取;红色和白色条形光源下的成像效果略有差异,白色较红色的图像亮度整体偏暗,得到的最适高度未随光源的变化而发生改变。试验结果具有一定普遍性,对于实际生产中图像处理和图像采集具有指导意义。
In the research process of using the image feature extraction method to identify the surface defects of sawn timber, the illumination of the light source in the vision system is one of the important factors to evaluate the image quality. In order to obtain better image data for identifying the surface defects of sawn timber, this study adopted a comprehensive evaluation method combining subjective and objective. Ratio, contrast, and image histogram from four angles were used to explore the response change of image quality. During the test, the light source of the vision system selected red LED strip lights and white LED strip lights. Through the combination of image quality evaluation method analysis and single-factor test verification, the vision system was tested under different light sensitivities, different light intensities, and different lighting conditions. An experimental study on the differences in imaging effects under different frequencies. The test results in this study showed that, although the image quality of the sawn timber surface was not significantly affected by the light intensity, the image quality of the sawn timber surface defects obtained under the condition of the light source at a working distance of 60 mm was better than other conditions. Along with the particularity of the black color, when the working distance of the light source of the vision system was at the high point and the low point, the image quality of the sawn timber surface dead knots did not change much, but the acquisition of the sawn timber surface defect image under the condition of 80 mm should be avoided. The imaging effect under the red LED strip light source and the white LED strip light source was slightly different. The brightness of the defect image under the white LED strip light source was darker than that under the red LED strip light source. The optimal working height of the system light source did not change with the change of light source intensity and frequency. The test results in this study have certain universality and practicability and have positive guiding significance for image processing and image acquisition in the actual production of wood processing and utilization.
作者
刁兴良
张伟
纪敏
王国富
DIAO Xingliang;ZHANG Wei;JI Min;WANG Guofu(Research Institute of Wood Industry,Chinese Academy of Forestry,Beijing 100091,China)
出处
《林业工程学报》
CSCD
北大核心
2022年第6期140-147,共8页
Journal of Forestry Engineering
基金
林业和草原科技成果国家级推广项目([2019]35号)
国家自然科学基金(31670721)。
关键词
锯材
节子
图像质量评价
信噪比
对比度
光源距离
sawn timber
knot
image quality assessment
signal-to-noise ratio
contrast ratio
light source distance
作者简介
刁兴良,男,研究方向为林业装备与信息化;通信作者:张伟,男,研究员。E-mail:wzhang@caf.ac.cn。
