[Wheat College] opencv tutorial function Summary

Opencv 2.4.4 contains 100 built-in examples.

Parter 1:
No1. adaptiveskindetector. cpp
Skin Detection Using the color information of the HSV space, the background cannot have too many colors similar to the skin color. The effect is not particularly good.
NO2. bagofwords_classification.cpp
A big string ...... I cannot understand it yet.
NO3. bgfg_codebook.cpp
Background separation. Enable the camera or read the video.
No4. bgfg_gmg.cpp
Cameras capture and separate the background Based on Motion.
No5. bgfg_segm.cpp
Gaussian processing video. Perform background segmentation before the tracking movement. Backgroundsubtractormog2 class.
No6. blobtrack_sample.cpp
Video Tracking. Tracks the moving objects in the video and uses green wireframes.
No7. brief_match_test.cpp
Use the brief descriptive operator to match two-dimensional image feature points. Line118 error .???
No8. build3dmodel. cpp
Create a 3D model. Model The Source Based on the given detector.
No9. calibration. cpp 3calibration. cpp
Out-of-camera calibration. According to the built-in function, corner points are extracted and then scaled, which results in poor performance.
No10. calibration_artificial
Automatically calibrate the camera according to the corner. After initialization, calibratecamera is used to find the angle and calibrate it. findchessboardcorners is used to estimate the effect.

How can this problem be solved.
No11. chamfer. cpp
Image matching. Search for the template image in the target image after the binary value of the image. It mainly calls the chamermatching function.
No12. ipvs. c
Profile search and acquisition. Cvfindcontours is a function.
No13. convert_cascade.c
Load the trained cascade classifier from the file or import it from the classifier database embedded in opencv and save it as a file.
No14. convexhull. cpp
Convex Hull. Calculate the convex hull after a random vertex is generated.
No15. cout_mat.cpp
Matrix output in opencv.
No16. ce-c delaunay2.cpp
The edge is located based on the random vertex and the cell structure of The KNN chart is calculated at the end.
No17. demhist. cpp
Histogram equalization is used to adjust the brightness and contrast of an image and output a black-and-white image.
No18. descriptor_extractor_matcher.cpp
7-8 parameters. Sift matching.
No19. detector_descriptor_evaluation.cpp
Calculate the detection operator. Various dataset.
No20. detector_descriptor_matcher_evaluation.cpp
Calculate detection operator matching. It is also a variety of dataset.
No21. DFT. cpp
Perform discrete Fourier transformation on the image. Mathematical transformation.
No22. distrans. cpp
Distance conversion. Calculates the distance between all non-zero elements in the input image and the nearest zero element.
No23. drawing. cpp
Simple painting points, lines, text, etc. Not explained.
No24. edge. cpp
Edge detection. Adjust the threshold through a slide bar and detect the image edge and display it with the help of a very simple code.
No25. em. cpp
Em clustering.
No26. fabmap_sample
Fab-mat match. Create a chow-Liu tree from the training data.
No27. facedetect. cpp smiledetect. cpp
Face detection. Based on the trained classifier, the facial image is detected and the detected facial features are marked with circle boxes or rectangular boxes of different colors.
No28. facerec_demo.cpp
Face recognition.
No29. fback. cpp fback_c.c
Calculate the optical flow of a video. The camera is turned on by default. Some cards are slow.
No30. filestorage. cpp
Mat matrix storage, read and write XML/yml files.
No31. find_obj.cpp
Examples of the surf algorithm. Search for objects in the sample image using matching.
No32. find_obj_calonder.cpp
Detects the target object through the training classification tree. Image training is required.
No33. find_obj_ferns.cpp
It is also the target detection. Quickly identifies key points based on random fern clusters.
No34. fitellipse. cpp
Elliptical fitting to search for image contour images. Findcontours is useful. The overall effect is not satisfactory.
No35. freak_demo.cpp
Use feature points for image matching. Feature description includes a. alahi, R. Ortiz, and P. vandregheynst. Freak: fast

Retina keypoint.
No36. gencolors. cpp
Enter the number of colors to generate a color stripe image. Color bandwidth 20.
No37. generic_descriptor_match.cpp
Surf image matching. The input parameters include two images and parameter data.
No38. houghlines. cpp houghcircles. cpp
Extract Straight lines or circles from the image using the Hough transform. The effect is normal. It is important that you use the system.
No39. image. cpp
Basic image and video reading, image noise and smooth processing.
No40. Kalman. cpp
Kalman filter. First, a motion model and an observation model are created. The algorithm results show the line between the estimated point and the actual point for one-dimensional point tracking of circular motion.
No41. kmeans. cpp
Cluster analysis. The K-means algorithm is used for clustering iteration after random points are generated on the plane. Because the cluster center is also randomly generated, the effect is poor.
No42. Laplace. cpp
It is also edge detection. The threshold value is adjusted by the slide bar. First, the image is filtered (Gaussian, mean, and median), and then the Laplace is used to detect the edge. The Sigma parameter is 3, which has the best effect.
No43. latentsvmdetect. cpp
Use latentsvm to detect the target.
No44. letter_recog.cpp
Demonstrate training different classifiers by using the UCI Character Library dataset.
No45. logpolar_bsm.cpp
Coordinate Conversion.
No46. matcher_simple.cpp
Surf image matching. There are few parameters, and the effect is similar to generic_descriptor_match.cpp.
No47. matching_to_many_images.cpp
Match multiple images. Powerful surf algorithm.
No48. meanshift_segmentation.cpp
Meanshift image segmentation. The three parameters spatialrad, colorrad, and maxpyrlevel are adjustable.
No49. minarea. cpp
After a random vertex is generated, the circle and rectangle with the smallest area of all vertices are calculated. Pure mathematical problems.
No50. morphology. c morphology2.cpp
Basic morphological operations, including on/off operations and expansion/corrosion operations.
No51. motempl. c
Motion Tracking.
No52. mser_sample.cpp
The mser method is used to extract the image contour. Maximally

Stable colour regions, mscr ).
No53. mushroom. cpp
Demonstrate creation of decision tree training using mushroom data
No54. one_way_sample.cpp
Feature Point Matching Based on Principal Component Analysis. It takes a long time to run ......
No55. opencv_version.cpp
The opencv version is displayed. A few simple lines of code.
No56. openexrimages_highdynamicrange_retina_tonemapping.cpp

Openexrimages_highdynamicrange_retina_tonemapping_video.cpp
Not clear.
No57. openni_capture.cpp
Open natural interaction video capturing. Depthgenerator.
No58. PCA. cpp
Main Component analysis algorithm. Reconstruction.
No59. peopledetect. cpp
Hog (histogram-of-oriented gradients) pedestrian or human detection, using hog features and SVM.
No60. phase_pai.cpp
Phase-Based Motion and azimuth tracking program for related images.
No61. points_classifier.cpp
Point category. Click the given point and class.
No62. polar_transform.c
Linear coordinates and polar coordinates convert each other. Images can be captured from cameras.
No63. pyramid_segmentation.c
Pyramid image segmentation.
No64. retinademo. cpp
Retina feature points detection.
No65. rgbdodometry. cpp
Visual ODPS algorithm. In order to estimate the rigid body transformation, we try to find the warping, that is, to maximize the different image scales of two consecutive rgbd frames.
No66. segment_objects.cpp
Video Tracking separates objects in motion.
No67. select3dobj. cpp
The collection of dataset objects and split masks shows how to use the camera's calibration mode. Calculates the plane on the calibration pattern of the single correspondence. Also show

Grabcut and so on.
No68. simpleflow_demo.cpp
An optical flow algorithm.
No69. squares. cpp
Find the rectangle.
No70. starter_imagelist.cpp
Read and display images based on the image list file yaml.
No71. starter_video.cpp
Open the video image, select the image, and save it as an image.
No72. stereo_calib.cpp
Camera stereo calibration.
No73. stereo_match.cpp
Stereo match.
No74. stitching. cpp stitching_detailed.cpp
Image stitching. It involves feature extraction, feature point matching, and image fusion. Stitcher class.
No75. tvl1_optical_flow.cpp
Optical Flow Video Tracking.
No76. tree_engine.cpp
Demonstrate the use of different decision trees cvdtree dtree; Decision Tree cvboost boost; boosted Tree Classifier supervised learning tree

Cvrtrees rtrees; Random Tree cvertrees ertrees; completely random tree.
No77. video_dmtx.cpp
Video.
No78. video_homography.cpp
Use features2d quick corner detection.
No79. videostab. cpp
Stable video.
No80. Watershed
Make Watershed Image segmentation.

Parter 2:
No1. camshiftdemo. cpp
Color target tracking. Follow the color spectrum of a region by clicking the mouse to track the video target.
NO2. connected_components.cpp
Connected area. Findcontours + drawcontours.
NO3. ipvs2.cpp
First draw a line chart and then detect the contour. Adjustable parameters.
No4. ffilldemo. cpp
Fill in the water. Search for points in the same color as the selected points in the image and mark them with different colors.
No5. grabcut. cpp
Split the image, select the rectangle frame with the mouse, extract the foreground, and separate the background. The results are quite good.
No6. hybridtrackingsample. cpp
Hybrid tracing. Hybridtracker encountered an error during debugging.
No7. imagelst_creator.cpp
Write the image name list in yaml or XML format.
No8. inpaint. cpp
Digital Image repair program, texture-based synthesis. Draw images randomly first, and press the "I" key to display the repaired images.
No9. linemod. cpp
Line196 error .???
No10. lkdemo. cpp
Point Tracking. Improved Lucas-kanade optical flow algorithm to detect Video moving targets. Click the target point to track the video.

Parter3:
No1. detection_based_tracker_sample.cpp
Examples used on Unix or android platforms. Detection-based tracking.

This is an example of all 100. If you have another idea about the 100 built-in examples of opencv, you can come to the it online education platform-wheat College to discuss with me.

 

[Wheat College] opencv tutorial function Summary