import parse_haar import sys if __name__ == '__main__': if len(sys.argv) < 3: print 'usage: %s cascade.xml outfile.txt' exit(1) in_filename, out_filename = sys.argv[1:3] cascade = parse_haar.parse_haar_xml(in_filename) with open(out_filename, 'w') as outf: for i, stage in enumerate(cascade.stages): outf.write('stage %i\n'%i) for j, feature in enumerate(stage.features): outf.write(' feature %i\n'%j) for k, shape in enumerate(feature.shapes): outf.write(' %s\n'%str(shape[0]))
parser.add_option('-m', '--multi_scale', action='store_true', dest='multi_scale', default=False, help='multiscale detection') (options, args) = parser.parse_args() (options, args) = parser.parse_args() if options.do_test: test() exit(0) cascade_filename = options.cascade src_dir = options.src_dir output_dir = options.output_dir use_multiscale = options.multi_scale haar_classifier = parse_haar.parse_haar_xml(cascade_filename) print str(haar_classifier) # get all images in dir image_filenames = [src_dir + os.path.sep + x for x in os.listdir(src_dir) if '.png' in x.lower()] # parameters scale_factor = 1.2 min_size = (40, 40) if not os.path.isdir(output_dir): print 'warning: creating outputdir %s'%output_dir os.mkdir(output_dir) for i, image_filename in enumerate(image_filenames): image = imageio.read(image_filename)