机器学习笔记 - 模式识别的应用场景之一简单车牌识别

一、关于模式识别

        模式（Pattern）是指具有某种特定性质的感知对象。一般情况下，待观察的事物都具有时空分布信息。模式识别（Pattern Recognition）又称为模式分类，指对待观察事物的各种信息进行处理、描述、分类和解释的过程。按照有无训练样本，模式识别可分为监督模式识别和非监督模式识别两种。

        模式识别的研究方向主要有两个：第一个是研究生物体是如何感知世界的；第二个是研究如何用机器（包括计算机）识别特定对象的模式。这些特定对象可以是字符、语音、图像等具体的事物，也可以是状态、程度、范围等抽象的表达。模式识别与数学、医学、心理学、语言学、物理学及计算机科学等都有关系。

机器学习笔记 - 简单了解模式识别在数据中寻找模式是一个有着悠久历史的问题。模式识别是一种数据分析方法，它使用机器学习 算法自动识别数据中的模式和规律。这些数据可以是任何东西，从文本和图像到声音或其他可量化的数据。模式识别系统可以快速准确地识别熟悉的模式。他们还可以识别和分类不熟悉的物体，从不同角度识别形状和物体，并识别图案和物体。2、模式识别的特点模式识别从数据中学习。即使部分可见，也能自动识别图案。应该能够识别熟悉的模式。应从不同的角度和形状识别图案。https://skydance.blog.csdn.net/article/details/126520391

        下面的代码都来自别人的github，是基于图像处理+支持向量机实现的，也是模式识别的范畴。比较典型也易于理解，是一个用于了解模式识别比较好的入门示例。 

二、车辆识别

        对于传入的车牌图像，基于旋转校正、阈值分割、边缘检测等常用的图像分割算法。颜色分割算法/边缘检测算法完成车牌区域定位；然后对车牌区域校正，分割字符区域；最后导入预训练的SVM分类器，识别字符区域，输出识别结果。

1、定位车牌

		if type(car_pic) == type(""):
			img = imreadex(car_pic)
		else:
			img = car_pic
		pic_hight, pic_width = img.shape[:2]
		if pic_width > MAX_WIDTH:
			pic_rate = MAX_WIDTH / pic_width
			img = cv2.resize(img, (MAX_WIDTH, int(pic_hight*pic_rate)), interpolation=cv2.INTER_LANCZOS4)
		
		if resize_rate != 1:
			img = cv2.resize(img, (int(pic_width*resize_rate), int(pic_hight*resize_rate)), interpolation=cv2.INTER_LANCZOS4)
			pic_hight, pic_width = img.shape[:2]
			
		print("h,w:", pic_hight, pic_width)
		blur = self.cfg["blur"]
		#高斯去噪
		if blur > 0:
			img = cv2.GaussianBlur(img, (blur, blur), 0)#图片分辨率调整
		oldimg = img
		img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
		#equ = cv2.equalizeHist(img)
		#img = np.hstack((img, equ))
		#去掉图像中不会是车牌的区域
		kernel = np.ones((20, 20), np.uint8)
		img_opening = cv2.morphologyEx(img, cv2.MORPH_OPEN, kernel)
		img_opening = cv2.addWeighted(img, 1, img_opening, -1, 0);

		#找到图像边缘
		ret, img_thresh = cv2.threshold(img_opening, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
		img_edge = cv2.Canny(img_thresh, 100, 200)
		#使用开运算和闭运算让图像边缘成为一个整体
		kernel = np.ones((self.cfg["morphologyr"], self.cfg["morphologyc"]), np.uint8)
		img_edge1 = cv2.morphologyEx(img_edge, cv2.MORPH_CLOSE, kernel)
		img_edge2 = cv2.morphologyEx(img_edge1, cv2.MORPH_OPEN, kernel)

		#查找图像边缘整体形成的矩形区域，可能有很多，车牌就在其中一个矩形区域中
		try:
			contours, hierarchy = cv2.findContours(img_edge2, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
		except ValueError:
			image, contours, hierarchy = cv2.findContours(img_edge2, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
		contours = [cnt for cnt in contours if cv2.contourArea(cnt) > Min_Area]
		print('len(contours)', len(contours))
		#一一排除不是车牌的矩形区域
		car_contours = []
		for cnt in contours:
			rect = cv2.minAreaRect(cnt)
			area_width, area_height = rect[1]
			if area_width < area_height:
				area_width, area_height = area_height, area_width
			wh_ratio = area_width / area_height
			#print(wh_ratio)
			#要求矩形区域长宽比在2到5.5之间，2到5.5是车牌的长宽比，其余的矩形排除
			if wh_ratio > 2 and wh_ratio < 5.5:
				car_contours.append(rect)
				box = cv2.boxPoints(rect)
				box = np.int0(box)
				#oldimg = cv2.drawContours(oldimg, [box], 0, (0, 0, 255), 2)
				#cv2.imshow("edge4", oldimg)
				#cv2.waitKey(0)

		print(len(car_contours))

		print("精确定位")
		card_imgs = []
		#矩形区域可能是倾斜的矩形，需要矫正，以便使用颜色定位
		for rect in car_contours:
			if rect[2] > -1 and rect[2] < 1:#创造角度，使得左、高、右、低拿到正确的值
				angle = 1
			else:
				angle = rect[2]
			rect = (rect[0], (rect[1][0]+5, rect[1][1]+5), angle)#扩大范围，避免车牌边缘被排除

			box = cv2.boxPoints(rect)
			heigth_point = right_point = [0, 0]
			left_point = low_point = [pic_width, pic_hight]
			for point in box:
				if left_point[0] > point[0]:
					left_point = point
				if low_point[1] > point[1]:
					low_point = point
				if heigth_point[1] < point[1]:
					heigth_point = point
				if right_point[0] < point[0]:
					right_point = point

			if left_point[1] <= right_point[1]:#正角度
				new_right_point = [right_point[0], heigth_point[1]]
				pts2 = np.float32([left_point, heigth_point, new_right_point])#字符只是高度需要改变
				pts1 = np.float32([left_point, heigth_point, right_point])
				M = cv2.getAffineTransform(pts1, pts2)
				dst = cv2.warpAffine(oldimg, M, (pic_width, pic_hight))
				point_limit(new_right_point)
				point_limit(heigth_point)
				point_limit(left_point)
				card_img = dst[int(left_point[1]):int(heigth_point[1]), int(left_point[0]):int(new_right_point[0])]
				card_imgs.append(card_img)
				#cv2.imshow("card", card_img)
				#cv2.waitKey(0)
			elif left_point[1] > right_point[1]:#负角度
				
				new_left_point = [left_point[0], heigth_point[1]]
				pts2 = np.float32([new_left_point, heigth_point, right_point])#字符只是高度需要改变
				pts1 = np.float32([left_point, heigth_point, right_point])
				M = cv2.getAffineTransform(pts1, pts2)
				dst = cv2.warpAffine(oldimg, M, (pic_width, pic_hight))
				point_limit(right_point)
				point_limit(heigth_point)
				point_limit(new_left_point)
				card_img = dst[int(right_point[1]):int(heigth_point[1]), int(new_left_point[0]):int(right_point[0])]
				card_imgs.append(card_img)
				#cv2.imshow("card", card_img)
				#cv2.waitKey(0)
		#开始使用颜色定位，排除不是车牌的矩形，目前只识别蓝、绿、黄车牌
		colors = []
		for card_index,card_img in enumerate(card_imgs):
			green = yello = blue = black = white = 0
			card_img_hsv = cv2.cvtColor(card_img, cv2.COLOR_BGR2HSV)
			#有转换失败的可能，原因来自于上面矫正矩形出错
			if card_img_hsv is None:
				continue
			row_num, col_num= card_img_hsv.shape[:2]
			card_img_count = row_num * col_num

			for i in range(row_num):
				for j in range(col_num):
					H = card_img_hsv.item(i, j, 0)
					S = card_img_hsv.item(i, j, 1)
					V = card_img_hsv.item(i, j, 2)
					if 11 < H <= 34 and S > 34:#图片分辨率调整
						yello += 1
					elif 35 < H <= 99 and S > 34:#图片分辨率调整
						green += 1
					elif 99 < H <= 124 and S > 34:#图片分辨率调整
						blue += 1
					
					if 0 < H <180 and 0 < S < 255 and 0 < V < 46:
						black += 1
					elif 0 < H <180 and 0 < S < 43 and 221 < V < 225:
						white += 1
			color = "no"

			limit1 = limit2 = 0
			if yello*2 >= card_img_count:
				color = "yello"
				limit1 = 11
				limit2 = 34#有的图片有色偏偏绿
			elif green*2 >= card_img_count:
				color = "green"
				limit1 = 35
				limit2 = 99
			elif blue*2 >= card_img_count:
				color = "blue"
				limit1 = 100
				limit2 = 124#有的图片有色偏偏紫
			elif black + white >= card_img_count*0.7:#TODO
				color = "bw"
			print(color)
			colors.append(color)
			print(blue, green, yello, black, white, card_img_count)
			#cv2.imshow("color", card_img)
			#cv2.waitKey(0)
			if limit1 == 0:
				continue
			#以上为确定车牌颜色
			#以下为根据车牌颜色再定位，缩小边缘非车牌边界
			xl, xr, yh, yl = self.accurate_place(card_img_hsv, limit1, limit2, color)
			if yl == yh and xl == xr:
				continue
			need_accurate = False
			if yl >= yh:
				yl = 0
				yh = row_num
				need_accurate = True
			if xl >= xr:
				xl = 0
				xr = col_num
				need_accurate = True
			card_imgs[card_index] = card_img[yl:yh, xl:xr] if color != "green" or yl < (yh-yl)//4 else card_img[yl-(yh-yl)//4:yh, xl:xr]
			if need_accurate:#可能x或y方向未缩小，需要再试一次
				card_img = card_imgs[card_index]
				card_img_hsv = cv2.cvtColor(card_img, cv2.COLOR_BGR2HSV)
				xl, xr, yh, yl = self.accurate_place(card_img_hsv, limit1, limit2, color)
				if yl == yh and xl == xr:
					continue
				if yl >= yh:
					yl = 0
					yh = row_num
				if xl >= xr:
					xl = 0
					xr = col_num
			card_imgs[card_index] = card_img[yl:yh, xl:xr] if color != "green" or yl < (yh-yl)//4 else card_img[yl-(yh-yl)//4:yh, xl:xr]
		#以上为车牌定位

2、识别字符

        车牌字符识别使用的算法是opencv的SVM， opencv的SVM使用代码来自于opencv附带的sample，StatModel类和SVM类都是sample中的代码。SVM训练使用的训练样本来自于github上的EasyPR的c++版本。由于训练样本有限，你测试时会发现，车牌字符识别，可能存在误差，尤其是第一个中文字符出现的误差概率较大。源码中有EasyPR中的训练样本，在train目录下，如果要重新训练请解压在当前目录下，并删除原始训练数据文件svm.dat和svmchinese.dat。

#以下为识别车牌中的字符
		predict_result = []
		roi = None
		card_color = None
		for i, color in enumerate(colors):
			if color in ("blue", "yello", "green"):
				card_img = card_imgs[i]
				gray_img = cv2.cvtColor(card_img, cv2.COLOR_BGR2GRAY)
				#黄、绿车牌字符比背景暗、与蓝车牌刚好相反，所以黄、绿车牌需要反向
				if color == "green" or color == "yello":
					gray_img = cv2.bitwise_not(gray_img)
				ret, gray_img = cv2.threshold(gray_img, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
				#查找水平直方图波峰
				x_histogram  = np.sum(gray_img, axis=1)
				x_min = np.min(x_histogram)
				x_average = np.sum(x_histogram)/x_histogram.shape[0]
				x_threshold = (x_min + x_average)/2
				wave_peaks = find_waves(x_threshold, x_histogram)
				if len(wave_peaks) == 0:
					print("peak less 0:")
					continue
				#认为水平方向，最大的波峰为车牌区域
				wave = max(wave_peaks, key=lambda x:x[1]-x[0])
				gray_img = gray_img[wave[0]:wave[1]]
				#查找垂直直方图波峰
				row_num, col_num= gray_img.shape[:2]
				#去掉车牌上下边缘1个像素，避免白边影响阈值判断
				gray_img = gray_img[1:row_num-1]
				y_histogram = np.sum(gray_img, axis=0)
				y_min = np.min(y_histogram)
				y_average = np.sum(y_histogram)/y_histogram.shape[0]
				y_threshold = (y_min + y_average)/5#U和0要求阈值偏小，否则U和0会被分成两半

				wave_peaks = find_waves(y_threshold, y_histogram)

				#for wave in wave_peaks:
				#	cv2.line(card_img, pt1=(wave[0], 5), pt2=(wave[1], 5), color=(0, 0, 255), thickness=2) 
				#车牌字符数应大于6
				if len(wave_peaks) <= 6:
					print("peak less 1:", len(wave_peaks))
					continue
				
				wave = max(wave_peaks, key=lambda x:x[1]-x[0])
				max_wave_dis = wave[1] - wave[0]
				#判断是否是左侧车牌边缘
				if wave_peaks[0][1] - wave_peaks[0][0] < max_wave_dis/3 and wave_peaks[0][0] == 0:
					wave_peaks.pop(0)
				
				#组合分离汉字
				cur_dis = 0
				for i,wave in enumerate(wave_peaks):
					if wave[1] - wave[0] + cur_dis > max_wave_dis * 0.6:
						break
					else:
						cur_dis += wave[1] - wave[0]
				if i > 0:
					wave = (wave_peaks[0][0], wave_peaks[i][1])
					wave_peaks = wave_peaks[i+1:]
					wave_peaks.insert(0, wave)
				
				#去除车牌上的分隔点
				point = wave_peaks[2]
				if point[1] - point[0] < max_wave_dis/3:
					point_img = gray_img[:,point[0]:point[1]]
					if np.mean(point_img) < 255/5:
						wave_peaks.pop(2)
				
				if len(wave_peaks) <= 6:
					print("peak less 2:", len(wave_peaks))
					continue
				part_cards = seperate_card(gray_img, wave_peaks)
				for i, part_card in enumerate(part_cards):
					#可能是固定车牌的铆钉
					if np.mean(part_card) < 255/5:
						print("a point")
						continue
					part_card_old = part_card
					#w = abs(part_card.shape[1] - SZ)//2
					w = part_card.shape[1] // 3
					part_card = cv2.copyMakeBorder(part_card, 0, 0, w, w, cv2.BORDER_CONSTANT, value = [0,0,0])
					part_card = cv2.resize(part_card, (SZ, SZ), interpolation=cv2.INTER_AREA)
					#cv2.imshow("part", part_card_old)
					#cv2.waitKey(0)
					#cv2.imwrite("u.jpg", part_card)
					#part_card = deskew(part_card)
					part_card = preprocess_hog([part_card])
					if i == 0:
						resp = self.modelchinese.predict(part_card)
						charactor = provinces[int(resp[0]) - PROVINCE_START]
					else:
						resp = self.model.predict(part_card)
						charactor = chr(resp[0])
					#判断最后一个数是否是车牌边缘，假设车牌边缘被认为是1
					if charactor == "1" and i == len(part_cards)-1:
						if part_card_old.shape[0]/part_card_old.shape[1] >= 8:#1太细，认为是边缘
							print(part_card_old.shape)
							continue
					predict_result.append(charactor)
				roi = card_img
				card_color = color
				break
				
		return predict_result, roi, card_color#识别到的字符、定位的车牌图像、车牌颜色

         识别结果如下

h,w: 394 631
len(contours) 3
2
精确定位
blue
3800 21 0 67 9 4641
no
1 420 897 24 270 20145
['粤', 'X', '8', '3', '3', '3', '8']

        车牌定位算法的参数受图像分辨率、色偏、车距影响，如果是基于深度学习的算法效果可能会更好。