编写python代码弓箭缺陷检测提取灰度特征，包含灰度归一化均值，灰度方差，并把缺陷位置画出来

可以使用 OpenCV 库实现弓箭缺陷检测并提取灰度特征。下面是样例代码：

import cv2
import numpy as np

# 读取图像
img = cv2.imread('arrow.jpg')

# 转为灰度图像
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

# 归一化
gray_norm = cv2.normalize(gray.astype('float'), None, 0.0, 1.0, cv2.NORM_MINMAX)

# 计算灰度均值和方差
mean, var = cv2.meanStdDev(gray_norm)
mean = float(mean)
var = float(var)

# Sobel 计算梯度
gx = cv2.Sobel(gray_norm, cv2.CV_32F, 1, 0, ksize=3)
gy = cv2.Sobel(gray_norm, cv2.CV_32F, 0, 1, ksize=3)
mag, ang = cv2.cartToPolar(gx, gy, angleInDegrees=True)

# 阈值分割
ret, thresh = cv2.threshold(mag, 0.2, 1.0, cv2.THRESH_BINARY)
thresh = thresh.astype('uint8')

# 查找轮廓
contours, hierarchy = cv2.findContours(thresh, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

# 绘制缺陷区域
for contour in contours:
    rect = cv2.boundingRect(contour)
    if rect[2] >= 10 and rect[3] >= 10: # 过滤掉过小的矩形
        cv2.rectangle(img, rect, (0, 0, 255), 2)

# 显示图像
cv2.imshow('img', img)
cv2.waitKey(0)
cv2.destroyAllWindows()

请注意，此处的缺陷检测方法仅供参考，具体应根据实际情况进行调整和优化。