人工智能：python+OpenCV实现视频跟踪流水线缺陷检测识别

from random import randint import cv2 import numpy as np import Defects as defects from matplotlib import pyplot as plt class Product: scratch_sum = 0 blot_sum = 0 count = 0 def __init__(self, p, xi, yi, x, y, w, h): # 工件ID self.pid = p # 中心坐标 self.centre_x = xi self.centre_y = yi # 边框 self.bound_x = x self.bound_y = y self.bound_w = w self.bound_h = h # 颜色 self.R = randint(0, 255) self.G = randint(0, 255) self.B = randint(0, 255) # 运动轨迹 self.tracks = [] # 0代表正常，1代表样品存在划痕，2代表样品存在污渍 self.state = 0 self.sample = 0 # 储存缺陷对象 self.defects = [] # 获取ID def getId(self): return self.pid def getBoundX(self): return self.bound_x def getBoundY(self): return self.bound_y def getBoundW(self): return self.bound_w def getBoundH(self): return self.bound_h def getX(self): # 获取中心x return self.centre_x def getY(self): # 获取中心y return self.centre_y # 更新位置 def updateCoords(self, xn, yn, x, y, w, h): self.centre_x = xn self.centre_y = yn self.bound_x = x self.bound_y = y self.bound_w = w self.bound_h = h # 增加轨迹位置 self.tracks.append([self.centre_x, self.centre_y]) # 获取轨迹 def getTracks(self): return self.tracks # 获取标记颜色 def getRGB(self): return self.R, self.G, self.B # 保存 def save_pic(self, frame): pic = frame[self.bound_y:self.bound_y + self.bound_h, self.bound_x:self.bound_x + self.bound_w] self.sample = pic # 改变图像大小 cv2.imwrite("./output/" + str(self.pid) + ".jpg", pic) # 缺陷识别和分类 def defect_detect(self): # 彩色图像转灰度图像 gray = cv2.cvtColor(self.sample, cv2.COLOR_BGR2GRAY) # 二值化处理(黑白图像) thresh = cv2.threshold(gray, 135, 250, cv2.THRESH_BINARY)[1] # 输出二值化处理的图像 cv2.imwrite("./output/" + str(self.pid) + "thresh.jpg", thresh) contour, hierar = cv2.findContours(thresh, cv2.RETR_TREE, cv2.CHAIN_APPROX_NONE) img = self.sample.copy() for cnt in contour: area = cv2.contourArea(cnt) if 100 < area <= 15000: x, y, w, h = cv2.boundingRect(cnt) ret = img[y:y + h, x: x + w] # 提取特征缺陷 gray = cv2.cvtColor(ret, cv2.COLOR_BGR2GRAY) thresh = cv2.threshold(gray, 135, 250, cv2.THRESH_BINARY_INV)[1] contours, hierarchy = cv2.findContours(thresh, cv2.RETR_TREE, cv2.CHAIN_APPROX_NONE) mask = np.zeros(gray.shape).astype(gray.dtype) cv2.fillPoly(mask, contours, (255, 255, 255))# 填充任意形状的多边形 cv2.imwrite("./output/" + str(self.pid) + "fillPoly.jpg", mask) result = cv2.bitwise_and(gray, mask) cv2.imwrite("./output/" + str(self.pid) + "bitwise_and.jpg", result)#得到缺陷部分 # 直方图计算 """ cv2.calcHist函数： imaes:输入的图像（） channels:选择图像的通道 mask:掩膜，是一个大小和image一样的np数组，其中把需要处理的部分指定为1，不需要处理的部分指定为0，一般设置为None，表示处理整幅图像 histSize:使用多少个bin(柱子)，一般为256 ranges:像素值的范围，一般为[0,255]表示0~255 """ hist = cv2.calcHist([result], [0], None, [256], [0, 255]) s = sum(hist) # 判断缺陷，计算各像素点值的个数百分比 for i in range(len(hist)): if hist[i] > 0: hist[i] = hist[i] / s # 根据规则来匹配各种缺陷的情况 hist_sum_scratch = 0 hist_sum_blot = 0 for i in range(90, 135): hist_sum_scratch = hist_sum_scratch + hist[i] for i in range(15, 90): hist_sum_blot = hist_sum_blot + hist[i] if hist_sum_scratch > 0.1: d = defects.Defect(1, x, y, w, h) self.defects.append(d) self.state = 1 # print("此处缺陷划痕") if hist_sum_blot > 0.1: d = defects.Defect(2, x, y, w, h) self.defects.append(d) self.state = 2 # print("此处缺陷污渍") if self.state == 1: Product.scratch_sum = Product.scratch_sum + 1 elif self.state == 2: Product.blot_sum = Product.blot_sum + 1 return self.defects