(资料图)
缺陷检测是一种通过计算机视觉技术来检测产品制造过程中的缺陷的方法。该技术可以检测出产品表面的缺陷,如裂纹、凹陷、划痕、气泡等,并且可以实时监测和诊断制造过程中的问题。在制造业中,机器视觉缺陷检测技术已经被广泛应用于各种产品的质量控制和检测工作中,如电子产品、汽车零部件、医疗器械等。
* 关闭窗口dev_close_window ()*关闭程序计数器,图形变量更新,窗口图形更新dev_update_off ()*设置图像路径Path := "lcd/mura_defects_blur_"*读取一张图像read_image (Image, Path + "01")*获取图像大小get_image_size (Image, Width, Height)*创建一个新窗体dev_open_window_fit_size (0, 0, Width, Height, 640, 480, WindowHandle)*设置字体信息:字体大小为16,字体为mono,粗体,斜体set_display_font (WindowHandle, 14, "mono", "true", "false")*设置区域的填充方式dev_set_draw ("margin")*设置线宽度dev_set_line_width (3)*设置输出对象的显示颜色dev_set_color ("red")ScaleFactor := 0.4*获取lines_gauss算子Sigma, Low, High三个参数值calculate_lines_gauss_parameters (17, [25,3], Sigma, Low, High)for f := 1 to 3 by 1 *读取一张图像 1=>"01" read_image (Image, Path + f$".2i") *获取彩色图像的R,G,B三分量 decompose3 (Image, R, G, B) * *将图像转化为频域图像 rft_generic (B, ImageFFT, "to_freq", "none", "complex", Width) *生成一个高斯滤波核 gen_gauss_filter (ImageGauss, 100, 100, 0, "n", "rft", Width, Height) *将频域图像核高斯滤波核进行卷积运算 convol_fft (ImageFFT, ImageGauss, ImageConvol) *将卷积后的图像转换为空间域图像 rft_generic (ImageConvol, ImageFFT1, "from_freq", "none", "byte", Width) *用缺陷图像减去背景图像(时域图像) sub_image (B, ImageFFT1, ImageSub, 2, 100) * 对上述图像进行抽点,变焦 zoom_image_factor (ImageSub, ImageZoomed, ScaleFactor, ScaleFactor, "constant") *获取变焦后的图像的ROI get_domain (ImageZoomed, Domain) *图像ROI进行腐蚀操作 erosion_rectangle1 (Domain, RegionErosion, 7, 7) *获取变焦图像中ROI区域内的图像 reduce_domain (ImageZoomed, RegionErosion, ImageReduced) *探测线和获取线宽度 lines_gauss (ImageReduced, Lines, Sigma, Low, High, "dark", "true", "gaussian", "true") *生成一个2D的齐次变换矩阵 hom_mat2d_identity (HomMat2DIdentity) *添加一个缩放因子到齐次变换矩阵 hom_mat2d_scale_local (HomMat2DIdentity, 1 / ScaleFactor, 1 / ScaleFactor, HomMat2DScale) *仿射变换 affine_trans_contour_xld (Lines, Defects, HomMat2DScale) * 显示图像 dev_display (Image) * 显示线缺陷 dev_display (Defects) stop () endfor
*关闭程序计数器,图形变量更新,窗口图形更新dev_update_off ()*关闭已打开的窗口dev_close_window ()************************************************第一步: 获取图像***************************************************读取图像read_image (Image, "image.png")*获取图像大小get_image_size (Image, Width, Height)*打开一个新窗口dev_open_window_fit_image (Image, 0, 0, Width, Width, WindowID)*设置字体信息:字体大小为16,字体为mono,粗体,斜体set_display_font (WindowID, 16, "mono", "true", "false")*设置区域的填充方式dev_set_draw ("margin")*设置线宽度dev_set_line_width (4)*显示图像dev_display (Image) ************************************************第二步: 分割图像*************************************************** 均值滤波,滤波核大小为7*7mean_image (Image, ImageMean, 7, 7)* 用局部阈值进行分割dyn_threshold (Image, ImageMean, DarkPixels, 5, "dark")* 联通处理connection (DarkPixels, ConnectedRegions)* 输出对象的显示颜色的数目dev_set_colored (12)dev_display (Image)dev_display (ConnectedRegions)************************************************第三步: 处理区域*************************************************** 过滤出缺陷区域select_shape (ConnectedRegions, SelectedRegions, "area", "and", 10, 1000)dev_display (Image)dev_display (SelectedRegions)* 将SelectedRegions中的所有区域合并union1 (SelectedRegions, RegionUnion)* 对合并区域进行膨胀处理,其结构元素为圆形,半径为3.5dilation_circle (RegionUnion, RegionDilation, 3.5)dev_display (Image)dev_display (RegionDilation)Message := "Region of the scratches after dilation"* 对膨胀后的区域进行求骨架skeleton (RegionDilation, Skeleton)* 对获取的骨架区域进行连通处理connection (Skeleton, Errors)dev_set_colored (12)dev_display (Image)dev_display (Errors)************************************************第三步: 获取大缺陷和小缺陷***************************************************过滤出大的缺陷select_shape (Errors, Scratches, "area", "and", 50, 10000)*过滤出小的缺陷select_shape (Errors, Dots, "area", "and", 1, 50)dev_display (Image)dev_set_color ("red")dev_display (Scratches)dev_set_color ("blue")dev_display (Dots)
*读取一张图像read_image (Image, "1.jpg") *将RGB图像转化为灰度图rgb1_to_gray (Image, GrayImage)*使用自动阈值分割图像前景区域binary_threshold (GrayImage, Region, "max_separability", "dark", UsedThreshold)*将前景区域中图像剪切出来reduce_domain (GrayImage, Region, ClipImage)*均值滤波mean_image (ClipImage, ImageMean, 200, 3) *局部阈值,分割出前景dyn_threshold (ClipImage, ImageMean, RegionDynThresh, 7, "light") *对分割的区域闭运算,填充间隙,平滑边界closing_circle (RegionDynThresh, RegionClosing, 3.5) *连通处理connection (RegionClosing, ConnectedRegions) *过滤出三角形区域select_shape (ConnectedRegions, SelectedRegions, "area", "and", 500, 2000) *填充三角形内部区域fill_up (SelectedRegions, RegionFillUp) *将三角形区域转换为凸度shape_trans (RegionFillUp, RegionTrans, "convex") *将区域转换为轮廓gen_contour_region_xld (RegionTrans, Contours, "border") *显示图像dev_display(GrayImage) *显示三角形区域dev_display(Contours)
* 检测毛刺ImageFiles := []ImageFiles[0] := "./fin1.png"ImageFiles[1] := "./fin2.png"ImageFiles[2] := "./fin3.png"for Index := 0 to |ImageFiles| - 1 by 1 read_image (Image, ImageFiles[Index]) binary_threshold (Image, Region, "max_separability", "light", UsedThreshold) closing_circle (Region, RegionClosing, 203.5) reduce_domain (Image, RegionClosing, ImageReduced) reduce_domain (Image, Region, ImageReduced1) *差分 difference (ImageReduced, ImageReduced1, RegionDifference) opening_circle (RegionDifference, RegionOpening, 3) dev_set_draw ("margin") dev_set_line_width (5) dev_display (Image) dev_display (RegionOpening) stop()endfor
*关闭程序计数器,图形变量,窗口图形更新dev_update_off() * Image Acquisition 01: Code generated by Image Acquisition 01ImageFiles := []ImageFiles[0] := "1.bmp"ImageFiles[1] := "2.bmp"for Index := 0 to |ImageFiles| - 1 by 1 *读取一张图像 read_image (Image, ImageFiles[Index]) *分割出环形区域 threshold (Image, Region, 100, 255) *对区域进行连通处理 connection (Region, ConnectedRegions) *过滤出圆环区域 select_shape (ConnectedRegions, SelectedConnectedRegions, ["area","circularity"], "and", [40000,0.6], [99999,1]) *将圆环区域中的图像剪切出来 reduce_domain (Image, SelectedConnectedRegions, ImageReduced) *将图像和高斯导数进行卷积运算 derivate_gauss (ImageReduced, DerivGaussImage, 1.5, "gradient") *分割出图像中的轮廓 threshold (DerivGaussImage, DerivGaussRegion, 9, 50) *对轮廓区域进行连通处理 connection (DerivGaussRegion, DerivGaussConnectedRegions) *获取的轮廓区域 select_shape_std (DerivGaussConnectedRegions, SelectedDerivGaussConnectedRegions, "max_area", 70) *填充该轮廓区域 fill_up (SelectedDerivGaussConnectedRegions, RegionFillUp) *求轮廓区域和填充的轮廓区域之间的补集区域 difference (RegionFillUp, SelectedDerivGaussConnectedRegions, RegionDifference) *对补集区域进行连通处理 connection (RegionDifference, ResultConnectedRegions) *获取缺陷区域 select_shape (ResultConnectedRegions, DefectResult, "area", "and", 50, 999) *显示图像 dev_display (Image) *显示缺陷区域 dev_display (DefectResult) stop()endfor
*读取一张图像 read_image (Image, "1.jpg") *获取图像大小 get_image_size(Image, Width, Height) *关闭已经打开的窗口 dev_close_window () *打开新窗口 dev_open_window(0, 0, Width, Height, "black", WindowHandle) //打开指定大小的窗口 *对图像进行阈值操作 threshold (Image, Region, 0, 50) *对区域进行连通处理 connection (Region, ConnectedRegions) *过滤出产品的外轮廓区域 select_shape (ConnectedRegions, SelectedRegions, "ra", "and", 10, 200) *对轮廓区域进行填充 fill_up (SelectedRegions, RegionFillUp) *将区域转化为最小外接圆 shape_trans (SelectedRegions, RegionTrans, "outer_circle") *通过补集运算获取产品缺口区域 difference (RegionTrans, RegionFillUp, RegionDifference) *对缺口区域进行腐蚀操作 erosion_circle (RegionDifference, RegionErosion, 1) *把一个区域转变为一个二进制字节图像。 region_to_bin (RegionErosion, BinImage, 255, 0, 656, 492) *对图像进行阈值操作 threshold (BinImage, Region1, 255, 255) *对阈值区域连通处理 connection (Region1, ConnectedRegions1) *设置输出对象显示颜色 dev_set_color("blue") *过滤出想要的缺口区域 select_shape (ConnectedRegions1, SelectedRegions1, ["ra","rb"], "and", [5,1], [10,5]) *统计出缺口的数目 count_obj (SelectedRegions1, Number) *显示图像 dev_display (Image) *显示缺陷区域 dev_display (SelectedRegions1)
*读取图像read_image (Image, "pcb")*关闭已经打开的窗口dev_close_window ()*获取图像大小get_image_size (Image, Width, Height)*打开新的窗口dev_open_window (0, 0, Width, Height, "black", WindowHandle)*显示图像dev_display (Image)*执行灰度开运算gray_opening_shape (Image, ImageOpening, 7, 7, "octagon")*执行灰度闭运算gray_closing_shape (Image, ImageClosing, 7, 7, "octagon")*局部阈值操作,分割出电路断线dyn_threshold (ImageOpening, ImageClosing, RegionDynThresh, 75, "not_equal")*显示图像dev_display (Image)*设置输出对象的显示颜色dev_set_color ("red")*设置区域的填充方式dev_set_draw ("margin")*显示电路断线区域dev_display (RegionDynThresh)
*读取一张图像read_image (Image, "1.png")*获取图像大小get_image_size (Image, Width, Height)*关闭已经打开的窗口dev_close_window ()*打开新窗口dev_open_window (0, 0, Width/2, Height/2, "black", WindowHandle)*对图像均值滤波,滤波核大小3*3mean_image(Image, ImageMean, 3, 3) *对图像均值滤波,滤波核大小12*12mean_image(Image, ImageMean2, 12, 12) *通过局部阈值分割出网格区域dyn_threshold(ImageMean, ImageMean2, RegionDynThresh, 5, "dark")*对网格区域进行连通处理connection(RegionDynThresh, ConnectedRegions)*过滤出网格区域select_shape(ConnectedRegions, SelectedRegions, "area", "and", 150, 50000) *开运算,获取网格竖线opening_rectangle1(SelectedRegions, RegionOpening1, 1, 10)*开运算,获取网格横线opening_rectangle1(SelectedRegions, RegionOpening2, 10, 1) *交集运算,获取横线和竖线的交点intersection(RegionOpening1, RegionOpening2, RegionIntersection)*对交点区域进行连通处理connection(RegionIntersection, ConnectedRegions1)*获取交点的坐标位置:行,列坐标area_center (ConnectedRegions1, Area, Row, Column)*显示图像dev_display (Image)*显示网格交点dev_display (ConnectedRegions1)
************************************************第一步: 程序初始化***************************************************关闭当前有打开的窗口dev_close_window ()*关闭程序计数器,图形变量,窗口图形更新dev_update_off ()*打开一个新窗口dev_open_window (0, 0, 640, 512, "black", WindowHandle)*设置字体信息:字体大小为16,字体为mono,粗体,斜体set_display_font (WindowHandle, 14, "mono", "true", "false")stop ()************************************************第二步: 批量读取图像***************************************************批量读取4张图像到Images图形变量里read_image (Images, "./shampoo_label_0" + [1:4])for I := 1 to 4 by 1 *从对象数组里选择对象元素 select_obj (Images, ObjectSelected, I) *显示选择的图像 dev_display (ObjectSelected) *延时0.5秒 wait_seconds (0.5)endfor************************************************第三步: 获取反照率和表面梯度*************************************************** 物像里的光照方向Tilts := [6.1,95.0,-176.1,-86.8]* 光源和物体之间的角度Slants := [41.4,42.6,41.7,40.9]ResultType := ["gradient","albedo"]photometric_stereo (Images, HeightField, Gradient, Albedo, Slants, Tilts, ResultType, "poisson", [], [])* 显示反射图dev_display (Albedo)stop ()* 使用梯度字段计算表面的高斯曲率,缺陷在曲率图像中通常易于检测derivate_vector_field (Gradient, MeanCurvature, 1.0, "mean_curvature")************************************************第四步: 获取检测到的缺陷*************************************************** 对曲率图像进行阈值threshold (MeanCurvature, Region, -10, -0.07)* 对区域进行开运算opening_circle (Region, RegionOpening, 1)* 对区域进行连通处理connection (RegionOpening, ConnectedRegions)* 将缺陷过滤出来select_shape (ConnectedRegions, Defects, "area", "and", 50, 99999)* 将区域转换为圆shape_trans (Defects, Circle, "outer_circle")* 设置区域填充方式dev_set_draw ("margin")* 设置输出对象的显示颜色dev_set_color ("red")* 显示反射图dev_display (Albedo)* 显示缺陷dev_display (Defects)* 显示缺陷外圆dev_display (Circle)
* 关闭窗口dev_close_window ()*关闭程序计数器,图形变量,窗口图形更新dev_update_off ()*图像路径Path := "mura_defects_texture_"*读取一张图像read_image (Image, Path + "01")*获取图像大小get_image_size (Image, Width, Height)*打开一个新窗口dev_open_window (0, 0, 640, 480, "black", WindowHandle)*设置字体信息:字体大小为16,字体为mono,粗体,斜体set_display_font (WindowHandle, 14, "mono", "true", "false")*设置区域的填充方式dev_set_draw ("margin")*设置线宽度dev_set_line_width (3)*设置输出对象显示颜色dev_set_color ("red")for F := 1 to 2 by 1 read_image (Image, Path + F$".2i") *获取彩色图像的三个通道,R,G,B decompose3 (Image, R, G, B) *缺陷是暗斑。 因此,通过原始图像减去背景照明,从而使缺陷变得更加明显。 estimate_background_illumination (B, ImageFFT1) *用缺陷图减去背景图 sub_image (B, ImageFFT1, ImageSub, 2, 100) *中值滤波 median_image (ImageSub, ImageMedian, "circle", 9, "mirrored") *提取分水岭盆地 watersheds_threshold (ImageMedian, Basins, 20) *计算灰度共生矩阵的灰度值特性,暗斑的能量非常低 cooc_feature_image (Basins, ImageMedian, 6, 0, Energy, Correlation, Homogeneity, Contrast) *Energy小于0.05,Mask为1,否则为0 Mask := Energy [<=] 0.05 select_mask_obj (Basins, Defects, Mask) * 显示图像 dev_display (Image) * 显示缺陷 dev_display (Defects) stop() endfor
************************************************第一 窗口初始化*****************************************************关闭已经打开的窗口dev_close_window () *关闭程序计数器,图形变量更新,窗口图形更新dev_update_off () *读取一张图像read_image (Image,"2.bmp") *获取图像大小get_image_size (Image, Width, Height) *打开新的窗口dev_open_window (0, 0, 640, 480, "black", WindowHandle) *设置区域填充方式dev_set_draw ("margin") *设置线宽度dev_set_line_width (2) *设置输出对象显示颜色数目dev_set_colored(12) ************************************************第二 图像预处理*****************************************************优化FFT的速度optimize_rft_speed (Width, Height, "standard") Sigma1 := 10.0 Sigma2 := 2.0 *构建高斯滤波器gen_gauss_filter (GsFilter1, Sigma1, Sigma1, 0.0, "none", "rft", Width, Height) *构建高斯滤波器gen_gauss_filter (GsFilter2, Sigma2, Sigma2, 0.0, "none", "rft", Width, Height)*两个滤波器相减sub_image (GsFilter1, GsFilter2, Filter, 1, 0) ************************************************第三 开始处理图像***************************************************** Image Acquisition 01: Code generated by Image Acquisition 01ImageFiles := []ImageFiles[0] := "1.bmp"ImageFiles[1] := "2.bmp"for Index := 0 to |ImageFiles| - 1 by 1 *读取一张图像read_image (Image, ImageFiles[Index])*将RGB彩色图转化为灰度图rgb1_to_gray (Image, Image) *对灰度图进行反选invert_image(Image, ImageInvert) *将图像转化为频域图像rft_generic (ImageInvert, ImageFFT, "to_freq", "sqrt", "complex", Width) *将频域图像和滤波核进行卷积运算convol_fft (ImageFFT, Filter, ImageConvol) *将频域图像转化为空间域图像rft_generic (ImageConvol, ImageFiltered, "from_freq", "n", "real", Width) *创建一个ROIgen_rectangle1(Rectangle,30,25,450,610) *将ROI区域的图像剪切reduce_domain(ImageFiltered, Rectangle, ROI) *中值滤波median_image(ROI, ImageMedian, "circle", 20, "mirrored") *使用高斯核进行图像平滑smooth_image (ROI, ImageSmooth, "gauss", 8) *通过阈值分割出缺陷区域threshold (ImageSmooth, Regions, -0.0016783, -0.0006434) *对缺陷区域进行连通处理connection(Regions, ConnectedRegions) *对区域进行膨胀操作dilation_circle (ConnectedRegions, RegionDilation, 5) *过滤出指定面积大小的缺陷select_shape (RegionDilation, SelectedRegions, "area", "and", 800, 99999) *根据区域生成XLD轮廓gen_contour_region_xld (SelectedRegions, Contours, "border") *将XLD拟合成圆fit_circle_contour_xld (Contours, "atukey", -1, 2, 0, 5, 2, Row, Column, Radius, StartPhi, EndPhi, PointOrder) *生成一个圆对象gen_circle_contour_xld (ContCircle, Row, Column, Radius+20, 0, 6.28318, "positive", 1)*显示图像dev_display (Image) *显示圆标记dev_display(ContCircle) stop() endfor
*关闭程序计数器,图形变量更新,窗口图形更新dev_update_window ("off")*读取一张图像read_image (Image, "plastic_mesh/plastic_mesh_01")*关闭打开的窗口dev_close_window ()*获取图像大小get_image_size (Image, Width, Height)*创建一个新窗口dev_open_window_fit_image (Image, 0, 0, Width, Height, WindowHandle)*设置字体信息:字体大小为16,字体为mono,粗体,斜体set_display_font (WindowHandle, 18, "mono", "true", "false")*设置区域的填充方式dev_set_draw ("margin")*设置线宽度dev_set_line_width (3)for J := 1 to 14 by 1 *读取一张图像 read_image (Image, "plastic_mesh/plastic_mesh_" + J$"02") *均值滤波 mean_image (Image, ImageMean, 49, 49) *局部阈值分割图像 dyn_threshold (Image, ImageMean, RegionDynThresh, 5, "dark") *对分割后的区域进行连通处理 connection (RegionDynThresh, ConnectedRegions) *过滤出缺陷网孔区域 select_shape (ConnectedRegions, ErrorRegions, "area", "and", 500, 99999) *统计出缺陷网孔的数目 count_obj (ErrorRegions, NumErrors) *显示图像 dev_display (Image) *设置对象的显示颜色 dev_set_color ("red") *显示缺陷网孔 dev_display (ErrorRegions) if (NumErrors > 0) *显示产品NG disp_message (WindowHandle, "Mesh not OK", "window", 24, 12, "black", "true") else *显示产品OK disp_message (WindowHandle, "Mesh OK", "window", 24, 12, "black", "true") endif stop() endfor
*****************************************************************************第一 图像预处理*********************************************************读取一张图像read_image(Image,"1.bmp")*对图像的灰度值在0到255范围内拉伸scale_image_max (Image, ImageScaleMax)*反选像素的位bit_not (ImageScaleMax, ImageNot)*生成一个椭圆元素gen_disc_se (SE, "byte", 45,45, 0)*黑帽运算,分割比临近暗的区域gray_bothat (ImageNot, SE, ImageBotHat)*生成ROI1gen_rectangle1 (ROI_0, 574.119, 268.867, 644.104, 337.728)*生成ROI2gen_rectangle1 (TMP_Region, 87.9029, 783.297, 179.989, 809.627)*合并两个ROIunion2 (ROI_0, TMP_Region, ROI_0)*生成ROI3gen_rectangle1 (TMP_Region, 743.558, 937.222, 839.327, 997.981)*合并ROIunion2 (ROI_0, TMP_Region, ROI_0)*将合并ROI区域的图像剪切reduce_domain (ImageBotHat, ROI_0, TemplateImage)*设置区域填充方式 dev_set_draw ("margin")*获取图像的数据,类型,高度,宽度get_image_pointer1 (TemplateImage, Pointer, Type, Width, Height)*显示图像dev_display (TemplateImage)*****************************************************************************第二 自动寻找最佳阈值********************************************************* 最大方差初始化为0MaxVariance := 0.0* 最佳分割灰度阈值从1遍历到255,初始阈值的选取可以取图像平均灰度值for ImgThreshold := 1 to 255 by 1 dev_display (TemplateImage) * 前景区域分割 threshold (TemplateImage, Region, ImgThreshold, 255) * 获得前景区域像素个数 area_center (Region, Area, Row, Column) * 获得前景区域均值和方差 intensity (Region,TemplateImage, Mean, Deviation) *对前景区域进行反选 complement (Region, RegionComplement) *获取背景区域的面积和坐标 area_center (RegionComplement, Area1, Row1, Column1) *获得背景区域像素个数、均值和方差 intensity (RegionComplement,TemplateImage, Mean1, Deviation1) * 计算类间方差 Otsu := Area*1.0/[Width*Height]*Area1*1.0/[Width*Height]*pow(Mean-Mean1,2) * 获取最大类间方差最佳阈值 if (Otsu>MaxVariance) MaxVariance := Otsu BestThreshold := ImgThreshold endifendfordev_display (TemplateImage)dev_set_color ("green")dev_set_draw ("fill") stop ()*****************************************************************************第三 分割缺陷目标********************************************************* 阈值操作,选取高亮目标,即为缺陷threshold (TemplateImage, Region1, BestThreshold, 255)* 对缺陷区域进行连通处理connection(Region1, ConnectedRegions)* 根据面积过滤出大的缺陷select_shape (ConnectedRegions, SelectedRegions1, "area", "and", 261.07, 1971)* 填充缺陷内部间隙fill_up(SelectedRegions1, RegionFillUp)* 显示图像dev_display(Image)* 显示缺陷区域dev_display(RegionFillUp)stop()
******************************************第一步 初始化***************************************************读取一张图像read_image (Hull, "1")*获取图像大小get_image_size (Hull, Width, Height)*关闭已经打开的窗口dev_close_window ()*打开一个新的窗口dev_open_window (0, 0, Width, Height, "black", WindowID)*显示图像dev_display (Hull) ******************************************第二步 图像处理***************************************************阈值操作,分割出吸嘴threshold (Hull, Dark, 0, 80)*补集运算,获取背景区域difference (Hull, Dark, Light)*对背景区域进行连通处理connection (Light, ConnectedRegions)*过滤出背景区域select_shape (ConnectedRegions, NoHullCand, "area", "and", 50000, 9999999)*对过滤的背景区域进行闭运算,填充背景间隙和平滑背景边界closing_circle (NoHullCand, NoHull, 13.5)*补集运算,获取吸嘴区域difference (Hull, NoHull, Region)*对吸嘴区域开运算opening_circle (Region, RegionOpening, 2.5)*对吸嘴区域进行连通处理connection (RegionOpening, ConnectedRegions)*过滤出吸嘴区域select_shape (ConnectedRegions, RegionHull, "area", "and", 5000, 9999999)*将吸嘴区域转换为凸包区域shape_trans (RegionHull, ConvexHull, "convex")*补集运算,获取吸嘴的缺口difference (ConvexHull, RegionHull, Region)*对吸嘴缺口区域进行连通处理connection (Region, ConnectedRegions)*过滤出吸嘴缺口select_shape (ConnectedRegions, LargeHoles, "area", "and", 2000, 99999)select_shape (LargeHoles, Holes, "convexity", "and", 0, 0.85)*显示图像dev_display (Hull)*设置输出对象的线宽度dev_set_line_width (5)*设置区域的填充方式dev_set_draw ("margin")*设置输出对象显示颜色为红色dev_set_color ("red")*显示吸嘴缺口dev_display (Holes)
*读入标准模板图片(图片名字为standard.bmp)read_image (Image, "./standard.bmp")*设置图像窗口get_image_size (Image, Width, Height)dev_close_window ()dev_open_window (0, 0, Width, Height, "black", WindowHandle)dev_display (Image)*选中模板区域(ROI)gen_rectangle1 (ROI_0, 201, 244, 385, 526)reduce_domain (Image, ROI_0, ImageReduced)*选中模板区域数字threshold (ImageReduced, Regions, 106, 255)connection (Regions, ConnectedRegions)select_shape (ConnectedRegions, SelectedRegions, "area", "and", 163.12, 10000)union1 (SelectedRegions, RegionUnion)*求模板区域坐标area_center (RegionUnion, Areastandard, Rowstandard, Columnstandard)*扣取模板区域图像,创建模板需要的是图像,不是区域reduce_domain (ImageReduced, RegionUnion, ImageReduced1)*创建模板create_shape_model (ImageReduced1, "auto", rad(0), rad(360), "auto", "auto", "use_polarity", "auto", "auto", ModelID)*读入图像路径,循环遍历每个图像list_files ("./", ["files","follow_links"], ImageFiles)tuple_regexp_select (ImageFiles, ["\\.(tif|tiff|gif|bmp|jpg|jpeg|jp2|png|pcx|pgm|ppm|pbm|xwd|ima|hobj)$","ignore_case"], ImageFiles)for Index := 0 to |ImageFiles| - 1 by 1 read_image (Image, ImageFiles[Index]) *模板匹配 find_shape_model (Image, ModelID, 0, rad(360), 0.5, 1, 0.5, "least_squares", 0, 0.9, Row, Column, Angle, Score) *判断匹配是否成功 if (|Score| > 0) *搜寻当前图像匹配成功的区域并求它的坐标和角度 dev_display_shape_matching_results (ModelID, "red", Row, Column, Angle, 1, 1, 0) *将模板区域仿射变换到匹配成功的区域 vector_angle_to_rigid (Rowstandard, Columnstandard, 0, Row, Column, Angle, HomMat2D) affine_trans_region (RegionUnion, RegionAffineTrans, HomMat2D, "nearest_neighbor") dev_display (Image) *blob方法提取当前图像的数字,并变成一个联通域 threshold (Image, Regions1, 106, 255) connection (Regions1, ConnectedRegions1) select_shape (ConnectedRegions1, SelectedRegions1, "area", "and", 163.12, 10000) union1 (SelectedRegions1, RegionUnion1) *仿射变换后的模板区域与当前区域求差集,即缺陷部分 difference (RegionAffineTrans, RegionUnion1, RegionDifference) *开运算去除很小点的噪声 opening_circle (RegionDifference, RegionOpening, 3) *求缺陷的面积 area_center (RegionOpening, Area, Row1, Column1) *缺陷面积大于阈值,说明有缺陷 if(Area>100) *设置缺陷区域的颜色 dev_set_color ("blue") *显示缺陷区域 dev_display (RegionOpening) *显示NG set_display_font (WindowHandle, 50, "mono", "true", "false") disp_message (WindowHandle,"NG", "window", 15, 40, "yellow", "false") *缺陷面积小于阈值,说明没缺陷 else *显示OK set_display_font (WindowHandle, 50, "mono", "true", "false") disp_message (WindowHandle,"OK", "window", 15, 40, "green", "false") endif *没有匹配到模板,直接NG else set_display_font (WindowHandle, 50, "mono", "true", "false") disp_message (WindowHandle,"NG", "window", 15, 40, "yellow", "false") endifendfor
标签:
缺陷检测是一种通过计算机视觉技术来检测产品制造过程中的缺陷的方法。该技术可以检测出产品表面的缺陷...
馅饼还是陷阱?百万豪车三折卖,博主:或面临“买得起用不起”问题
EurekaClient是一种用于向EurekaServer注册服务的客户端,它将应用程序的信息发送给EurekaServer,并定期向EurekaSe
1、不少用户将自己的电脑升级成Win10系统后发现开机变得很慢,动不动就要30几秒,让人烦躁。2、那么有没...
近日雪龙集团披露,截至2023年3月31日公司股东户数为1 14万户,较12月31日减少200 0户,减幅为1 73%...
期待已久的手游中土指挥官即将登陆九游,这款手机游戏吸引了大批玩家的关注,想下载这款游戏,有很多粉...
智通财经APP获悉,4月7日,上交所发布沪市监管概览。4月3日至4月7日,上交所对*ST泽达(688555 SH)、*ST...
4月7日,嘉亨家化公告显示,公司第二届第八次董事会会议于2023年4月7日以现场与通讯相结合的方式召开,...
周口日报全媒体记者任富强通讯员任君箫何强仲春时节,新绿初绽,这里生机盎然。“用好身边、眼前的红色...
利好消息加持下,国际原油价格上演底部反转,近半个月已涨超20%。继以沙特为首的OPEC+成员国集体宣布减...
4月10日,由21世纪经济报道举办的首届“公募REITs宣传周”即将拉开帷幕。经过十几年的不懈探索,我国公...
联想今天预热了首款小新mini高性能主机,宣布将于4月28日发布。
1、难以找到,怎么着都找不到。2、目前很没有办法,基本上是采用毒杀、诱捕等方法消灭。3、生物防治法可...
文 乐居财经研究院孙肃博3月结束,在港上市的内资房企陆续交出2022年答卷。乐居财经研究院重磅推出2022...
世界杯冠军阿根廷排名第1,这是他们时隔六年重返世界第一。中国男足的世界排名下滑一位,降至世界第81,...
”4月1日上午,东华大学80余名二年级学生来到广富林郊野公园内的哈哈农场,学习如何为梨树疏花。这是学...
中指研究院报道:根据中房指数系统百城价格指数,2023年03月,淮安新建住宅价格环比下跌0 07%,样本平...
美国一名工作人员牵着警犬开展工作(资料图)海外网4月7日电据美国福克斯新闻网6日报道,一名美国密歇根...
一、仲裁具备哪些优势仲裁具备的优势如下:1 自愿性。提交仲裁须双方当事人自愿,达成仲裁协议,当事人...
美菲2014年达成《强化防务合作协议》,该协议允许美军进入和使用美菲双方同意的“军事地点”,并在这些...
X 关闭
X 关闭