什么是圖像分割 圖像分割(Image Segmentation)是圖像處理最重要的處理手段之一 圖像分割的目標是將圖像中像素根據一定的規則分為若干(N)個cluster集合，每個集合包含一類像素。 根據算法分為監督學習方法和無監督學習方法，圖像分割的算法多數都是無監督學習方法 - KMeans 距離變換常見算法有兩種 - 不斷膨脹/腐蝕得到 - 基於倒角距離 分水嶺變換常見的算法 - 基於浸泡理論實現
cv::distanceTransform( InputArray src, OutputArray dst, OutputArray labels, //離散維諾圖輸出 int distanceType, // DIST_L1/DIST_L2, int maskSize, // 3x3,最新的支持5x5，推薦3x3、 int labelType=DIST_LABEL_CCOMP //dst輸出8位或者32位的浮點數，單一通道，大小與輸入圖像一致 ) cv::watershed( InputArray image, InputOutputArray markers )
處理流程 1. 將白色背景變成黑色-目的是為后面的變換做准備 2. 使用filter2D與拉普拉斯算子實現圖像對比度提高，sharp 3. 轉為二值圖像通過threshold 4. 距離變換 5. 對距離變換結果進行歸一化到[0~1]之間 6. 使用閾值，再次二值化，得到標記 7. 腐蝕得到每個Peak - erode 8. 發現輪廓 – findContours 9. 繪制輪廓- drawContours 10. 分水嶺變換 watershed 11. 對每個分割區域着色輸出結果
int main(int argc, char** argv) { char input_win[] = "input image"; char watershed_win[] = "watershed segmentation demo"; Mat src = imread(STRPAHT2); if (src.empty()) { printf("could not load image...\n"); return -1; } namedWindow(input_win, CV_WINDOW_AUTOSIZE); imshow(input_win, src); // 將白色背景變成黑色-為后面的變換做准備 for (int row = 0; row < src.rows; row++) { for (int col = 0; col < src.cols; col++) { if (src.at<Vec3b>(row, col) == Vec3b(255, 255, 255)) { src.at<Vec3b>(row, col)[0] = 0; src.at<Vec3b>(row, col)[1] = 0; src.at<Vec3b>(row, col)[2] = 0; } } } //namedWindow("black background", CV_WINDOW_AUTOSIZE); //imshow("black background", src); // sharpen Mat kernel = (Mat_<float>(3, 3) << 1, 1, 1, 1, -8, 1, 1, 1, 1); Mat imgLaplance; Mat sharpenImg = src; //使用filter2D與拉普拉斯算子實現圖像對比度提高，sharp filter2D(src, imgLaplance, CV_32F, kernel, Point(-1, -1), 0, BORDER_DEFAULT); src.convertTo(sharpenImg, CV_32F); Mat resultImg = sharpenImg - imgLaplance; resultImg.convertTo(resultImg, CV_8UC3); imgLaplance.convertTo(imgLaplance, CV_8UC3); imshow("sharpen image", resultImg); // convert to binary Mat binaryImg; cvtColor(src, resultImg, CV_BGR2GRAY); // 轉為二值圖像通過threshold threshold(resultImg, binaryImg, 40, 255, THRESH_BINARY | THRESH_OTSU); imshow("binary image", binaryImg); Mat distImg; // 每一個非零點距離離自己最近的零點的距離 distanceTransform(binaryImg, distImg, DIST_L1, CV_DIST_C, 5); // 歸一化 normalize(distImg, distImg, 0, 1, NORM_MINMAX); imshow("distance result", distImg); // 使用閾值，再次二值化，得到標記 threshold(distImg, distImg, .4, 1, THRESH_BINARY); Mat k1 = Mat::ones(13, 13, CV_8UC1); // 膨脹/腐蝕 erode(distImg, distImg, k1, Point(-1, -1)); imshow("distance binary image", distImg); // markers Mat dist_8u; distImg.convertTo(dist_8u, CV_8U); vector<vector<Point>> contours; // 發現輪廓 findContours(dist_8u, contours, RETR_EXTERNAL, CHAIN_APPROX_SIMPLE, Point(0, 0)); // 繪制輪廓 Mat markers = Mat::zeros(src.size(), CV_32SC1); for (size_t i = 0; i < contours.size(); i++) { drawContours(markers, contours, static_cast<int>(i), Scalar::all(static_cast<int>(i) + 1), -1); } circle(markers, Point(5, 5), 3, Scalar(255, 255, 255), -1); imshow("my markers", markers * 1000); // 分水嶺變換 watershed(src, markers); Mat mark = Mat::zeros(markers.size(), CV_8UC1); markers.convertTo(mark, CV_8UC1); bitwise_not(mark, mark, Mat()); imshow("watershed image", mark); // 對每個分割區域着色輸出結果 vector<Vec3b> colors; for (size_t i = 0; i < contours.size(); i++) { int r = theRNG().uniform(0, 255); int g = theRNG().uniform(0, 255); int b = theRNG().uniform(0, 255); colors.push_back(Vec3b((uchar)b, (uchar)g, (uchar)r)); } Mat dst = Mat::zeros(markers.size(), CV_8UC3); for (int row = 0; row < markers.rows; row++) { for (int col = 0; col < markers.cols; col++) { int index = markers.at<int>(row, col); if (index > 0 && index <= static_cast<int>(contours.size())) { dst.at<Vec3b>(row, col) = colors[index - 1]; } else { dst.at<Vec3b>(row, col) = Vec3b(0, 0, 0); } } } imshow("Final Result", dst); waitKey(0); return 0; }