阴影环境下，如何生成适配7段数码管数字提取的二值图？

正在开展7段数码管数字提取项目，已成功提取LED显示屏的ROI，但在阴影环境下难以生成合适的二值图像，无法用cv2.findContours准确识别数字。现有处理后的二值图效果不佳，数字的部分笔画丢失。

原始照片：

提取后的黑白照片：

现有代码：

img_name = 'test2.jpeg'
image = cv2.imread(img_name)

image = imutils.resize(image, height=1000)
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
blurred = cv2.GaussianBlur(gray, (5, 5), 0)
edged = cv2.Canny(blurred, 50, 200, 255)

#cv2.imshow("test", edged)
#cv2.waitKey(0)

cnts = cv2.findContours(edged.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cnts = imutils.grab_contours(cnts)
cnts = sorted(cnts, key=cv2.contourArea, reverse=True)

displayCnt = None
# loop over the contours
for c in cnts:
    # approximate the contour
    peri = cv2.arcLength(c, True)
    approx = cv2.approxPolyDP(c, 0.02 * peri, True)
    # if the contour has four vertices, then we have found
    # the thermostat display
    if len(approx) == 4:
        displayCnt = approx
        break
warped = four_point_transform(gray, displayCnt.reshape(4, 2))
output = four_point_transform(image, displayCnt.reshape(4, 2))

thresh = cv2.threshold(warped, 222, 255, cv2.THRESH_BINARY_INV+cv2.THRESH_OTSU)[1]
cv2.imwrite("black.png", thresh)

阴影导致图像局部亮度差异大，全局阈值（如OTSU）易失效，可尝试以下方法：

1. 自适应阈值二值化 #

基于局部区域计算阈值，适配明暗不均场景，替换原阈值代码：

# blockSize取奇数，C为阈值调整常数，可根据效果微调
thresh = cv2.adaptiveThreshold(warped, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, 
                               cv2.THRESH_BINARY_INV, 11, 2)

2. 光照补偿+阈值处理 #

先提升局部对比度，再做阈值：

• 直方图均衡化：

equalized = cv2.equalizeHist(warped)
thresh = cv2.threshold(equalized, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[1]

• CLAHE（限制对比度自适应均衡化，避免噪声过度增强）：

clahe = cv2.createCLAHE(clipLimit=2.0, tileGridSize=(8,8))
equalized = clahe.apply(warped)
thresh = cv2.threshold(equalized, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[1]

3. 颜色通道分离处理 #

若数字为特定颜色（如红色），提取对应通道处理更抗干扰：

import numpy as np
# 基于彩色ROI（对应代码中的output）处理
hsv = cv2.cvtColor(output, cv2.COLOR_BGR2HSV)
# 红色HSV范围，需根据实际场景微调
lower_red1 = np.array([0, 120, 70])
upper_red1 = np.array([10, 255, 255])
lower_red2 = np.array([170, 120, 70])
upper_red2 = np.array([180, 255, 255])
mask = cv2.inRange(hsv, lower_red1, upper_red1) | cv2.inRange(hsv, lower_red2, upper_red2)
# 形态学操作修复细节
kernel = np.ones((3,3), np.uint8)
thresh = cv2.morphologyEx(mask, cv2.MORPH_CLOSE, kernel)

4. 形态学优化 #

二值化后用膨胀+腐蚀修复断裂笔画：

kernel = np.ones((2,2), np.uint8)
thresh = cv2.dilate(thresh, kernel, iterations=1)
thresh = cv2.erode(thresh, kernel, iterations=1)

内容的提问来源于stack exchange，提问作者Chak Wing Mak