def text_detect(img,
MAX_HORIZONTAL_GAP=30,
MIN_V_OVERLAPS=0.6,
MIN_SIZE_SIM=0.6,
TEXT_PROPOSALS_MIN_SCORE=0.7,
TEXT_PROPOSALS_NMS_THRESH=0.3,
TEXT_LINE_NMS_THRESH=0.3,
MIN_RATIO=1.0,
LINE_MIN_SCORE=0.8,
TEXT_PROPOSALS_WIDTH=5,
MIN_NUM_PROPOSALS=1,
textmodel='darknet_detect'):
if textmodel == 'darknet_detect':
boxes, scores = darknet_detect.text_detect(np.array(img))
else:
boxes, scores = opencv_dnn_detect.text_detect(np.array(img))
boxes = np.array(boxes, dtype=np.float32)
scores = np.array(scores, dtype=np.float32)
textdetector = TextDetector(MAX_HORIZONTAL_GAP, MIN_V_OVERLAPS,
MIN_SIZE_SIM)
shape = img.size[::-1]
boxes = textdetector.detect(boxes, scores[:, np.newaxis], shape,
TEXT_PROPOSALS_MIN_SCORE,
TEXT_PROPOSALS_NMS_THRESH,
TEXT_LINE_NMS_THRESH, MIN_RATIO,
LINE_MIN_SCORE, TEXT_PROPOSALS_WIDTH,
MIN_NUM_PROPOSALS)
text_recs, tmp = get_boxes(np.array(img), boxes)
newBox = []
rx = 1
ry = 1
for box in text_recs:
x1, y1 = (box[0], box[1])
x2, y2 = (box[2], box[3])
x3, y3 = (box[6], box[7])
x4, y4 = (box[4], box[5])
newBox.append([
x1 * rx, y1 * ry, x2 * rx, y2 * ry, x3 * rx, y3 * ry, x4 * rx,
y4 * ry
])
return newBox, tmp
def text_detect(img,
MAX_HORIZONTAL_GAP=70,
MIN_V_OVERLAPS=0.6,
MIN_SIZE_SIM=0.6,
TEXT_PROPOSALS_MIN_SCORE=0.15,
TEXT_PROPOSALS_NMS_THRESH=0.3):
# 文本框
boxes, scores = detect.text_detect(np.array(img), thresh=TEXT_PROPOSALS_MIN_SCORE) # 耗时的地方
boxes = np.array(boxes, dtype=np.float32)
scores = np.array(scores, dtype=np.float32)
textdetector = TextDetector(MAX_HORIZONTAL_GAP, MIN_V_OVERLAPS, MIN_SIZE_SIM)
shape = img.size[::-1]
# 对textNet检测出来的框进行进一步的过滤，nms
boxes = textdetector.detect(boxes,
scores[:, np.newaxis], # 变为列
shape,
TEXT_PROPOSALS_NMS_THRESH,
)
# 画框结束
text_recs, tmp = draw_boxes(im=np.array(img), bboxes=boxes)
return text_recs, tmp