如何基于OpenCV按列统计图像中字符矩形框的数量？

原始字符图像：

已实现单个字符矩形框选的代码：

img = cv2.imread('/content/drive/MyDrive/project/t2.jpg')
image = cv2.resize(img,None,None,0.4,0.4)
#cv2_imshow(image)

gray = cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)
ret,thresh = cv2.threshold(gray,190,255,cv2.THRESH_BINARY_INV)
kernel = np.ones((1,1), np.uint8)
img_dilation = cv2.dilate(thresh, kernel, iterations=1)
ctrs,_= cv2.findContours(img_dilation.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
sorted_ctrs = sorted(ctrs, key=lambda ctr: cv2.boundingRect(ctr)[0])
for i, ctr in enumerate(sorted_ctrs):
    x, y, w, h = cv2.boundingRect(ctr)
    roi = image[y:y+h, x:x+w]
    cv2.rectangle(image,(x,y),( x + w, y + h ),(36,255,12),2)
    if w >2  and h > 2:
      #os.chdir('{}'.format(folder_out))
      roi = cv2.resize(roi,(224,224))
      #cv2.imwrite('letter{}.jpg'.format(i), roi)
cv2_imshow(image)

当前框选结果：

期望实现按列统计矩形框数量，目标效果如下：

简易实现代码 #

import cv2
import numpy as np

img = cv2.imread('/content/drive/MyDrive/project/t2.jpg')
image = cv2.resize(img, None, None, 0.4, 0.4)
result_img = image.copy()

gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
ret, thresh = cv2.threshold(gray, 190, 255, cv2.THRESH_BINARY_INV)
kernel = np.ones((1,1), np.uint8)
img_dilation = cv2.dilate(thresh, kernel, iterations=1)
ctrs, _ = cv2.findContours(img_dilation.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
sorted_ctrs = sorted(ctrs, key=lambda ctr: cv2.boundingRect(ctr)[0])

# 收集有效矩形框的中心x坐标及位置信息
valid_rects = []
for ctr in sorted_ctrs:
    x, y, w, h = cv2.boundingRect(ctr)
    if w > 2 and h > 2:
        center_x = x + w // 2
        valid_rects.append((center_x, x, y, w, h))

# 按列分组：设定阈值判断同列
col_threshold = 15  # 可根据图像实际情况调整
columns = []
for rect in valid_rects:
    cx = rect[0]
    placed = False
    for col in columns:
        avg_cx = np.mean([r[0] for r in col])
        if abs(cx - avg_cx) < col_threshold:
            col.append(rect)
            placed = True
            break
    if not placed:
        columns.append([rect])

# 按列的左x坐标排序，保证从左到右顺序
columns.sort(key=lambda col: col[0][1])

# 绘制框选和列统计数
for col_idx, col in enumerate(columns):
    col_count = len(col)
    col_left_x = col[0][1]
    # 绘制当前列所有矩形框
    for rect in col:
        x, y, w, h = rect[1], rect[2], rect[3], rect[4]
        cv2.rectangle(result_img, (x, y), (x+w, y+h), (36,255,12), 2)
    # 在列上方标注数量
    cv2.putText(result_img, str(col_count), (col_left_x, 30), 
                cv2.FONT_HERSHEY_SIMPLEX, 1, (0,0,255), 2)

cv2_imshow(result_img)
# 控制台输出各列数量
print("各列字符数：", [len(col) for col in columns])

核心逻辑说明 #

1. 过滤无效轮廓：先筛掉宽高小于2的小轮廓，避免噪声干扰统计
2. 同列判断：用矩形框的中心x坐标均值做参考，设定阈值（col_threshold），中心x差值小于阈值则归为同一列
3. 排序与标注：对列按左x坐标排序后，在列的上方用红色字体标注该列的矩形框数量

内容的提问来源于stack exchange，提问作者Nara Nart