{"id":1044,"date":"2015-09-22T11:31:48","date_gmt":"2015-09-22T18:31:48","guid":{"rendered":"http:\/\/visualgdb.com\/w\/?p=1044"},"modified":"2015-09-30T11:16:17","modified_gmt":"2015-09-30T18:16:17","slug":"using-opencv-library-on-raspberry-pi","status":"publish","type":"post","link":"https:\/\/visualgdb.com\/tutorials\/raspberry\/opencv\/","title":{"rendered":"Using OpenCV library on Raspberry Pi"},"content":{"rendered":"

This tutorial shows how to use the OpenCV 2 library in cross-compiled projects for Raspberry Pi 2. Before you begin, download the Raspberry Pi Jessie image<\/a> and write it to the SD card. Then get a cross-toolchain compatible with the image from the gnutoolchains.com<\/a> website.<\/p>\n

    \n
  1. Connect to your Raspberry Pi via SSH and run the “sudo apt-get install libopencv-dev” command to install OpenCV:\"01-inst\"<\/a><\/li>\n
  2. Start Visual Studio and open the VisualGDB Linux Project Wizard: \"02-linuxprj\"<\/a><\/li>\n
  3. OpenCV comes with configuration files for CMake that simplify including the library from CMake projects. Hence we will use CMake to build our demo application. Select it in the wizard:\"03-newcmake\"<\/a><\/li>\n
  4. Select “Build the project with a cross-compiler” and choose your Raspberry Pi board in the “deployment computer” field:\"04-cross\"<\/a><\/li>\n
  5. Press “Finish” to create your project. Then open VisualGDB Project Properties, go to the CMake page and click “Synchronize sysroot”. Ensure that \/opt\/vc<\/strong> and \/usr\/share\/OpenCV<\/strong> (not<\/strong><\/span> lowercase \/usr\/share\/opencv<\/strong>) are selected:\"05-resync\"<\/a><\/li>\n
  6. Open the CMakeLists.txt file in the newly created project and add the following line after add_executable():\n
    find_package(OpenCV REQUIRED)<\/pre>\n
    Also add ${OpenCV_LIBS} to target_libraries() statement: \"06-findpkg\"<\/a><\/li>\n
  7. Open VisualGDB Project Properties and set the CMAKE_PREFIX_PATH variable for the CMake command to point at <sysroot>\\usr\\share\\OpenCV:\"07-prefix\"<\/a><\/li>\n
  8. Replace the contents of the main source file with the following code from the find_contours sample in OpenCV:\n
    \/**\r\n\u00a0* @function findContours_Demo.cpp\r\n\u00a0* @brief Demo code to find contours in an image\r\n\u00a0* @author OpenCV team\r\n\u00a0*\/\r\n\r\n#include \"opencv2\/highgui\/highgui.hpp\"\r\n#include \"opencv2\/imgproc\/imgproc.hpp\"\r\n#include <iostream>\r\n#include <stdio.h>\r\n#include <stdlib.h>\r\n\r\nusing namespace cv;\r\nusing namespace std;\r\n\r\nMat src; Mat src_gray;\r\nint thresh = 100;\r\nint max_thresh = 255;\r\nRNG rng(12345);\r\n\r\n\/\/\/ Function header\r\nvoid thresh_callback(int, void*);\r\n\r\n\/**\r\n\u00a0* @function main\r\n\u00a0*\/\r\nint main(int, char** argv)\r\n{\r\n\u00a0 \/\/\/ Load source image and convert it to gray\r\n\u00a0\u00a0 \u00a0src = imread(argv[1], 1);\r\n\r\n\u00a0\u00a0 \u00a0\u00a0 \/\/\/ Convert image to gray and blur it\r\n\u00a0\u00a0 \u00a0cvtColor(src, src_gray, COLOR_BGR2GRAY);\r\n\u00a0\u00a0 \u00a0blur(src_gray, src_gray, Size(3, 3));\r\n\r\n\u00a0\u00a0 \u00a0\u00a0 \/\/\/ Create Window\r\n\u00a0\u00a0 \u00a0const char* source_window = \"Source\";\r\n\u00a0\u00a0 \u00a0namedWindow(source_window, WINDOW_AUTOSIZE);\r\n\u00a0\u00a0 \u00a0imshow(source_window, src);\r\n\r\n\u00a0\u00a0 \u00a0createTrackbar(\" Canny thresh:\", \"Source\", &thresh, max_thresh, thresh_callback);\r\n\u00a0\u00a0 \u00a0thresh_callback(0, 0);\r\n\r\n\u00a0\u00a0 \u00a0waitKey(0);\r\n\u00a0\u00a0 \u00a0return (0);\r\n}\r\n\r\n\/**\r\n\u00a0* @function thresh_callback\r\n\u00a0*\/\r\nvoid thresh_callback(int, void*)\r\n{\r\n\u00a0\u00a0 \u00a0Mat canny_output;\r\n\u00a0\u00a0 \u00a0vector<vector<Point> > contours;\r\n\u00a0\u00a0 \u00a0vector<Vec4i> hierarchy;\r\n\r\n\u00a0\u00a0 \u00a0\u00a0 \/\/\/ Detect edges using canny\r\n\u00a0\u00a0 \u00a0Canny(src_gray, canny_output, thresh, thresh * 2, 3);\r\n\u00a0\u00a0 \u00a0\/\/\/ Find contours\r\n\u00a0\u00a0 \u00a0findContours(canny_output, contours, hierarchy, CV_RETR_TREE, CV_CHAIN_APPROX_SIMPLE, Point(0, 0));\r\n\r\n\u00a0\u00a0 \u00a0\u00a0 \/\/\/ Draw contours\r\n\u00a0\u00a0 \u00a0Mat drawing = Mat::zeros(canny_output.size(), CV_8UC3);\r\n\u00a0\u00a0 \u00a0for (size_t i = 0; i < contours.size(); i++)\r\n\u00a0\u00a0 \u00a0{\r\n\u00a0\u00a0 \u00a0\u00a0\u00a0 \u00a0Scalar color = Scalar(rng.uniform(0, 255), rng.uniform(0, 255), rng.uniform(0, 255));\r\n\u00a0\u00a0 \u00a0\u00a0\u00a0 \u00a0drawContours(drawing, contours, (int)i, color, 2, 8, hierarchy, 0, Point());\r\n\u00a0\u00a0 \u00a0}\r\n\r\n\u00a0\u00a0 \u00a0\u00a0 \/\/\/ Show in a window\r\n\u00a0\u00a0 \u00a0namedWindow(\"Contours\", WINDOW_AUTOSIZE);\r\n\u00a0\u00a0 \u00a0imshow(\"Contours\", drawing);\r\n}<\/pre>\n<\/li>\n
  9. Upload any PNG image that will be used to check out contour detection to the Raspberry Pi.\u00a0 In this tutorial we will use the Lena.png file from the OpenCV package:\"18-lena\"<\/a><\/li>\n
  10. Specify the path to the uploaded image in the command-line arguments to your application via VisualGDB Project Properties:\"08-args\"<\/a><\/li>\n
  11. Press F5 to start debugging the application. You will see two windows: one with the original image and another one with the contours:\"09-windows\"<\/a>You can move the trackbar in the left window to change the threshold value for the edge detection algorithm.<\/li>\n<\/ol>\n
    To learn how to cross-compile OpenCV 3 for Raspberry Pi and step through its source code, follow this tutorial<\/a>.<\/p>\n
    Read this tutorial<\/a> for detailed instructions on using OpenCV with the Raspberry Pi camera module.<\/p>\n","protected":false},"excerpt":{"rendered":"
    This tutorial shows how to use the OpenCV 2 library in cross-compiled projects for Raspberry Pi 2. Before you begin,<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[12],"tags":[33,85],"_links":{"self":[{"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/posts\/1044"}],"collection":[{"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/comments?post=1044"}],"version-history":[{"count":6,"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/posts\/1044\/revisions"}],"predecessor-version":[{"id":1087,"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/posts\/1044\/revisions\/1087"}],"wp:attachment":[{"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/media?parent=1044"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/categories?post=1044"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/visualgdb.com\/w\/wp-json\/wp\/v2\/tags?post=1044"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}