17鼠标响应与操作
•回调函数参数: int event, int x, int y, int flags, void ** userdata* •Event 表示鼠标事件
•(x, y) 表示当前鼠标位置
•Flags 表示鼠标状态
•Userdata 表示回调用户数据，可以为空
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 b1 = cv.imread("123.jpg") img = np.copy(b1) x1 = -1 y1 = -1 x2 = -1 y2 = -1 def mouse_drawing(event, x, y, flags, param): global x1, y1, x2, y2 if event == cv.EVENT_LBUTTONDOWN: x1 = x y1 = y if event == cv.EVENT_MOUSEMOVE: if x1 <0 or y1 <0: return x2 = x y2 = y dx = x2 - x1 dy = y2 - y1 if dx > 0 and dy > 0: b1[:,:,:] = img[:,:,:] cv.rectangle(b1, (x1, y1), (x2, y2), (0, 0, 255), 2, 8, 0) if event == cv.EVENT_LBUTTONUP: x2 = x y2 = y dx = x2 - x1 dy = y2 - y1 if dx > 0 and dy > 0: b1[:,:,:] = img[:,:,:] cv.rectangle(b1, (x1, y1), (x2, y2), (0, 0, 255), 2, 8, 0) x1 = -1 x2 = -1 y1 = -1 y2 = -1 def mouse_demo(): cv.namedWindow("mouse_demo",cv.WINDOW_AUTOSIZE) cv.setMouseCallback("mouse_demo",mouse_drawing) while True: cv.imshow("mouse_demo",b1) c = cv.waitKey(10) if c == 27: break cv.destroyAllWindows()
18图像像素类型转换与归一化
归一化函数 •cv.normalize( src, dst[, alpha[, beta[, norm_type[, dtype[, mask]]]]] ) -> dst
•src表示输入图像, dst表示输出图像
•alpha, beta 默认是1， 0，是归一化的区间值
•norm_type默认是NORM_L2,
•norm_type常用是NORM_MINMAX
Imread读入默认是uint8, 转换为float32,通过imshow显示之前，必须归一化到[0~1]之间。
把float32的归一化图像转换为uint8类型：np.uint8(image*255)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 #归一化 def trackbar_callback(pos): print(pos) def norm_demo(): image_unit8 = cv.imread("123.jpg") cv.imshow("image_uint8",image_unit8) img_f32 = np.float32(image_unit8) # cv.imshow("img_f32",img_f32) # cv.normalize(img_f32, img_f32, 1, 0, cv.NORM_MINMAX) # cv.imshow("norm-img_f32",img_f32) # cv.waitKey(0) # cv.destroyAllWindows() cv.namedWindow("norm_demo",cv.WINDOW_AUTOSIZE) cv.createTrackbar("mormtype", "norm_demo", 10, 3, trackbar_callback) while True: dst = np.float32(image_unit8) pos = cv.getTrackbarPos("normtype","norm-demo") if pos == 0: cv.normalize(dst, dst, 1, 0, cv.NORM_MINMAX) if pos == 1: cv.normalize(dst, dst, 1, 0, cv.NORM_L1) if pos == 2: cv.normalize(dst, dst, 1, 0, cv.NORM_L2) if pos == 3: cv.normalize(dst, dst, 1, 0, cv.NORM_INF) cv.imshow("norm-demo",img_f32) c = cv.waitKey(50) if c == 27: break cv.destroyAllWindows()
19图像几何变换 •cv.warpAffine(src, M, dsize[, dst[, flags[, borderMode[, borderValue]]]] ) -> dst
•src表示输入图像
•M 表示2x3变换矩阵
•dsize表示目标图像dst的大小
•支持平移变换、放缩变换、旋转变换
获取旋转矩阵 •旋转矩阵获取cv.getRotationMatrix2D
•Center表示旋转中心, angle表示度数，大于零表示逆时针旋转, scale表示放缩尺度大小。
翻转与特殊角度旋转 •cv.flip(src, flipCode[, dst] ) ->dst
•cv.rotate(src, rotateCode[, dst] ) -> dst
•src表示输入图像
•flipCode支持0水平、1垂直，-1对角线翻转，
•rotateCode支持旋转90°，180°，270°
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 #图像几何变换 def affine_demo(): image = cv.imread("123.jpg") h, w, c = image.shape cx = int(w / 2) cy = int(h / 2) cv.imshow("image",image) M = np.zeros((2,3), dtype=np.float32) M[0, 0] = .7 M[1, 1] = .7 M[0, 2] = 0 M[1, 2] = 0 print("M(2x3) = \n", M) dst = cv.warpAffine(image, M, (int(w*.7), int(h*.7))) cv.imshow("rescale-demo",dst) cv.imwrite("result.png",dst) #获取旋转矩阵，degree > 0 表示逆时针旋转，原点在左上角 M = cv.getRotationMatrix2D((w/2, h/2), 45.0, 1.0) dst = cv.warpAffine(image, M, (w,h)) cv.imshow("rotate-demo",dst) dst = cv.flip(image, 0) cv.imshow("flip-demo",dst) cv.waitKey(0) cv.destroyAllWindows()
20视频读写处理 视频标准与格式 •SD(Standard Definition)标清480P
•HD(High Definition)高清720P/1080P
•UHD(Ultra High Definition)超高清4K/2160P
•分辨率表示
•SD-640x480, 704x480, 720x480, 848x480等
•HD-960x720,1280x720,1440x1080,1920x1080
•UHD-4K,2160P
视频读取函数 cv.VideoCapture ( filename, index, apiPreference)
•filename表示视频文件
•Index表示USB摄像头或者web camera的索引
•apiPreference = CAP_ANY意思自动决定第三方视频库如： cv.CAP_FFMPEG， cv.CAP_DSHOW
查询视频属性 •VideoCaput的get方法
•cv.CAP_PROP_FRAME_WIDT
•cv.CAP_PROP_FRAME_HEIGHT
•cv.CAP_PROP_FPS（对视频流来说是0）
•cv.CAP_PROP_FOURCC
•cv.CAP_PROP_FRAME_COUNT
视频文件保存 cv.VideoWriter(
filename, 保存文件名称
fourcc, 编码方式
fps, 帧率
frameSize 视频帧大小，与实现大小相符 ，否则无法保存
[, isColor] )
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 #20视频读写处理 def video_demo(): cap = cv.VideoCapture("456.avi") # query video file metadata fps = cap.get(cv.CAP_PROP_FPS) frame_w = cap.get(cv.CAP_PROP_FRAME_WIDTH) frame_h = cap.get(cv.CAP_PROP_FRAME_HEIGHT) print(fps, frame_w, frame_h) # encode mode # fourcc =cv.VideoWriter_fourcc(*'vp09') fourcc = cap.get(cv.CAP_PROP_FOURCC) # create Video writer writer = cv.VideoWriter('output.mp4', int(fourcc), fps, (int(frame_w), int(frame_h))) # loop read frame until last frame while True: ret, frame = cap.read() if ret is not True: break hsv = cv.cvtColor(frame, cv.COLOR_BGR2HSV) cv.imshow("hsv",hsv) cv.imshow("frame",frame) c = cv.waitKey(1) if c == 27: break writer.write(frame) # release camera resource cap.release() writer.release()
