You cannot select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
40 lines
1.1 KiB
Python
40 lines
1.1 KiB
Python
# 导入cv相关库
|
|
import cv2
|
|
import numpy as np
|
|
from PIL import ImageFont
|
|
from PIL import Image
|
|
from PIL import ImageDraw
|
|
# 导入依赖包
|
|
import hyperlpr3 as lpr3
|
|
|
|
|
|
def draw_plate_on_image(img, box, text, font):
|
|
x1, y1, x2, y2 = box
|
|
cv2.rectangle(img, (x1, y1), (x2, y2), (139, 139, 102), 2, cv2.LINE_AA)
|
|
cv2.rectangle(img, (x1, y1 - 20), (x2, y1), (139, 139, 102), -1)
|
|
data = Image.fromarray(img)
|
|
draw = ImageDraw.Draw(data)
|
|
draw.text((x1 + 5, y1 - 20), text, (255, 255, 255), font=font)
|
|
res = np.asarray(data)
|
|
|
|
return res
|
|
|
|
|
|
# 中文字体加载
|
|
font_ch = ImageFont.truetype("../resource/font/platech.ttf", 20, 0)
|
|
|
|
# 实例化识别对象
|
|
catcher = lpr3.LicensePlateCatcher(detect_level=lpr3.DETECT_LEVEL_HIGH)
|
|
# 读取图片
|
|
image = cv2.imread("/Users/tunm/Downloads/tlj.webp")
|
|
|
|
# 执行识别算法
|
|
results = catcher(image)
|
|
for code, confidence, type_idx, box in results:
|
|
# 解析数据并绘制
|
|
text = f"{code} - {confidence:.2f}"
|
|
image = draw_plate_on_image(image, box, text, font=font_ch)
|
|
|
|
# 显示检测结果
|
|
cv2.imshow("w", image)
|
|
cv2.waitKey(0) |