Opencv项目实战：23 物体计数和表单信息.zip

import cv2 import numpy as np import os import pandas as pd class VideoCap(): """ 自定义的摄像头调取类 """ def CapInit(self,mode=0,w=640,h=480,l=150): """ 摄像头初始化函数 :param mode: 0-web,1-ex camera :param w: 窗口宽度，默认为640 :param h: 窗口长度，默认为480 :param l: 亮度，默认为150 """ self.cap = cv2.VideoCapture(mode) self.mode=mode self.cap.set(3, w) self.cap.set(4, h) self.cap.set(10, l) def read(self): """ 摄像头读取函数 :return: 仅仅返回图片 """ _, img = self.cap.read() return img def readPath(folder_path): image_files = [] for filename in os.listdir(folder_path): if filename.endswith(".png") or filename.endswith(".jpg"): image_files.append(os.path.join(folder_path, filename)) return image_files def ImagePath(recongnitionMode): if len(recongnitionMode)>10: save_dir = "./Pic/img" excel_path = "./Pic/data.xlsx" return save_dir, excel_path else: save_dir = "./Video/img" excel_path = "./Video/data.xlsx" return save_dir, excel_path def Bbox_img(img, bbox_scale=0.5, bbox_color=(0, 0, 255)): """ 默认在图像中央显示红框 :param img: 输入图像 :param bbox_scale: 占图像的比例，默认为50% :param bbox_color: 边框颜色 :return: 返回含有边框的图像 """ height, width = img.shape[:2] box_width = int(width * bbox_scale) box_height = int(height * bbox_scale) start_x = int((width - box_width) / 2) start_y = int((height - box_height) / 2) end_x = start_x + box_width end_y = start_y + box_height postion=[(start_x, start_y),(end_x, end_y)] img_with_box = img.copy() cv2.rectangle(img_with_box, postion[0], postion[1], bbox_color, 2) return postion,img_with_box def maskBbox(img, position): """ 创建遮罩,对框外的图像进行高斯模糊处理 :param img: 输入图像 :param position: Bbox_img的返回值 :return: 受过遮罩保护的图像，对非遮罩区域进行高斯模糊处理 """ mask = np.zeros_like(img) cv2.rectangle(mask, position[0], position[1], (255, 255, 255), -1) blurred_img = cv2.GaussianBlur(img, (57, 57), 0) img_with_blur = np.where(mask == 0, blurred_img, img) return img_with_blur def onTrackbarChange(scale_value): """ 轨迹栏函数，主要用于全局的定义bbox_scale, brightness_factor """ global bbox_scale, brightness_factor bbox_scale = scale_value / 1000.0 + 0.1 brightness_factor = scale_value / 100.0 + 0.5 def empty(a): """ 轨迹栏函数 """ pass def preProcessing(img): """ 预处理阶段 :param img: 图像 :return: 经过了高斯、canny，膨胀、闭运算操作 """ imgPre = cv2.GaussianBlur(img, (5, 5), 3) thresh1 = cv2.getTrackbarPos("Threshold1", "Settings") thresh2 = cv2.getTrackbarPos("Threshold2", "Settings") imgPre = cv2.Canny(imgPre, thresh1, thresh2) kernel = np.ones((3, 3), np.uint8) imgPre = cv2.dilate(imgPre, kernel, iterations=1) imgPre = cv2.morphologyEx(imgPre, cv2.MORPH_CLOSE, kernel) return imgPre def findContours(img, imgPre, minArea=1000, sort=True, filter=0, c=(255, 0, 0)): """ 寻找物体的边缘信息，采用框选的方式，对于物体的表现更加的准确，经过测试对于轮廓的检测效果很差 :param img: 图片 :param imgPre: 预处理后的图片 :param minArea: 最小面积阈值，用于过滤小面积的边缘。默认值为 1000 :param sort: 是否按照面积进行排序。如果为 True，按照面积从大到小排序。默认值为 True :param filter: 过滤多边形的边数。如果为 0，则不进行过滤；如果为其他正整数，则只保留边数等于该值的多边形。默认值为 0 :param c: 边框和点的颜色，默认为红色 :return: """ conFound = [] imgContours = img.copy() contours, hierarchy = cv2.findContours(imgPre, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE) total = 0 for cnt in contours: area = cv2.contourArea(cnt) if area > minArea: peri = cv2.arcLength(cnt, True) approx = cv2.approxPolyDP(cnt, 0.02 * peri, True) if len(approx) == filter or filter == 0: x, y, w, h = cv2.boundingRect(approx) cx, cy = x + (w // 2), y + (h // 2) if 3000<= area <= 16000: cv2.rectangle(imgContours, (x, y), (x + w, y + h), c, 2) cv2.circle(imgContours, (x + (w // 2), y + (h // 2)), 5, c, cv2.FILLED) conFound.append({"cnt": cnt, "area": area, "bbox": [x, y, w, h], "center": [cx, cy]}) total+=1 if sort: conFound = sorted(conFound, key=lambda x: x["area"], reverse=True) return imgContours, conFound, total def stackImages(scale,imgArray): """ :param scale: 规格 :param imgArray: 图片形式列表 :return: 堆叠后的图片 """ rows = len(imgArray) cols = len(imgArray[0]) rowsAvailable = isinstance(imgArray[0], list) width = imgArray[0][0].shape[1] height = imgArray[0][0].shape[0] if rowsAvailable: for x in range ( 0, rows): for y in range(0, cols): if imgArray[x][y].shape[:2] == imgArray[0][0].shape [:2]: imgArray[x][y] = cv2.resize(imgArray[x][y], (0, 0), None, scale, scale) else: imgArray[x][y] = cv2.resize(imgArray[x][y], (imgArray[0][0].shape[1], imgArray[0][0].shape[0]), None, scale, scale) if len(imgArray[x][y].shape) == 2: imgArray[x][y]= cv2.cvtColor( imgArray[x][y], cv2.COLOR_GRAY2BGR) imageBlank = np.zeros((height, width, 3), np.uint8) hor = [imageBlank]*rows for x in range(0, rows): hor[x] = np.hstack(imgArray[x]) ver = np.vstack(hor) else: for x in range(0, rows): if imgArray[x].shape[:2] == imgArray[0].shape[:2]: imgArray[x] = cv2.resize(imgArray[x], (0, 0), None, scale, scale) else: imgArray[x] = cv2.resize(imgArray[x], (imgArray[0].shape[1], imgArray[0].shape[0]), None,scale, scale) if len(imgArray[x].shape) == 2: imgArray[x] = cv2.cvtColor(imgArray[x], cv2.COLOR_GRAY2BGR) hor= np.hstack(imgArray) ver = hor return ver def Adjusted_image(img,brightness_factor = 1.5): """ :param img: 图片 :param brightness_factor: 亮度调整因子 :return: """ image_float = img.astype(np.float32) adjusted_image = image_float * brightness_factor adjusted_image = np.clip(adjusted_image, 0, 255) adjusted_image = adjusted_image.astype(np.uint8) return adjusted_image def excelmation(excel_path,data): """ :param excel_path: 电子表格路径 :param data: 写入数据类型 :return: """ if not os.path.exists(excel_path): df = pd.DataFrame(columns=["Time", "Image Name", "Total"]) df.to_excel(excel_path, index=False, sheet_name="Data") df = pd.DataFrame(data, columns=["Time", "Image Name", "Total"]) df.to_excel(excel_path, index=False, sheet_name="Data") def drawContour(img, imgPre, minArea=15, show=True): """ :param img: 图片 :param imgPre: 预处理图片 :param minArea: 最小面积 :return: """ imgContours, conFound, total = findContours(img, imgPre, minArea) imgStacked = stackImages(1, [img, imgContours]) cv2.putText(imgStacked, f"Count: {tot