panel_segmentation.panel_detection.PanelDetection.plotEdgeAz

PanelDetection.plotEdgeAz(test_results, no_lines=5, no_figs=1, save_img_file_path=None, plot_show=False)[source]

This function is used to generate plots of the image with its azimuth It can generate three figures or one. For three figures, that include the input image, the hough transform space and the input images with detected lines. For single image, it only outputs the input image with detected lines.

Parameters

  • test_results (nparray float64 or unit8) – 8-bit input image. This variable represents the predicted images from the segmentation model. Hence the dimension must be [a,b,c,d] where [a] is the number of images, [b,c] are the dimensions of the image - 640 x 640 in this case and [d] is 3 - RGB

  • no_lines (int) – default is 10. This variable tells the function the number of dominant lines it should examine.

  • no_figs (int) – 1 or 3. If the number of figs is 1, it outputs the mask with Hough lines and the predicted azimuth. However, if the number of lines is 3, it gives three plots.

    1. The input image,

    2. Hough transform search space

    3. Unput image with houghlines and the predicted azimuth

  • save_img_file_path (string) – You can pass as input the location to save the plots

  • plot_show (boolean) – If False, it will supress the plot as an output and just save the plots in a folder

Returns

  • Figure

  • Plot of the masked image, with detected Hough Lines and azimuth

  • estimate.