This repository has been archived by the owner on Jan 20, 2024. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 42
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
[0.5.1] release ncc find_template && opencv orb.
- Loading branch information
1 parent
da43acd
commit 70b4e4e
Showing
7 changed files
with
155 additions
and
138 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,146 @@ | ||
| #ifndef cv_orb_hpp | ||
| #define cv_orb_hpp | ||
|
|
||
| #include "maix_image.h" | ||
|
|
||
| /* | ||
| from maix import * | ||
| orb = image.orb() | ||
| print(orb) | ||
| src = None | ||
| for i in range(10): | ||
| src = camera.capture() | ||
| display.show(src) | ||
| while True: | ||
| img = camera.capture() | ||
| res = orb.match(src, img, limit=30, max=100, dump=0) | ||
| # print(len(res)) | ||
| for point in res: | ||
| # print(point) | ||
| img.draw_circle(point[3], point[4], 4, color=(0, 255, 0), thickness=1) | ||
| display.show(img) | ||
| */ | ||
| struct cv_orb | ||
| { | ||
| cv::Ptr<cv::ORB> detector; | ||
| std::vector<cv::KeyPoint> queryPoints, trainPoints; | ||
| cv::Mat queryDescriptors, trainDescriptors; | ||
| void *cache_src; | ||
|
|
||
| cv_orb(int nfeatures = 500, float scaleFactor = 1.2f, int nlevels = 8, int edgeThreshold = 31, int firstLevel = 0, int WTA_K = 2, int scoreType = 0, int patchSize = 31, int fastThreshold = 20) | ||
| { | ||
| // printf("cv_orb: %d %f %d %d %d %d %d %d %d\n", nfeatures, scaleFactor, nlevels, edgeThreshold, firstLevel, WTA_K, scoreType, patchSize, fastThreshold); | ||
| this->detector = cv::ORB::create(nfeatures, scaleFactor, nlevels, edgeThreshold, firstLevel, WTA_K, (cv::ORB::ScoreType)scoreType, patchSize, fastThreshold); | ||
| this->cache_src = NULL; | ||
| } | ||
|
|
||
| py::list match(maix_image &src, maix_image &dst, int limit = 10, float min = 0.5f, float max = 100.0f, bool crossCheck = false, bool dump = 0) | ||
| { | ||
|
|
||
| py::list ret; | ||
| if (NULL == src._img || NULL == dst._img) | ||
| return ret; | ||
|
|
||
| if (src._img->width != dst._img->width && src._img->height != dst._img->height) | ||
| { | ||
| printf("[orb] match images of the same size are required\r\n"); | ||
| return ret; | ||
| } | ||
|
|
||
| cv::Mat img_src(src._img->height, src._img->width, CV_8UC3, src._img->data); | ||
| if (this->cache_src != src._img) | ||
| { | ||
| cv::cvtColor(img_src, img_src, cv::COLOR_RGB2GRAY); | ||
| this->detector->detectAndCompute(img_src, cv::Mat(), queryPoints, queryDescriptors); | ||
| // printf("queryPoints: %d\n", queryPoints.size()); | ||
| if (queryPoints.size() != 0) | ||
| { | ||
| this->cache_src = src._img; | ||
| } | ||
| else | ||
| { | ||
| this->cache_src = NULL; | ||
| printf("[orb] this image get queryPoints.size() == 0, replace it with another image.\r\n"); | ||
| return ret; | ||
| } | ||
| } | ||
|
|
||
| // printf("queryDescriptors: rows %d\n", queryDescriptors.rows); | ||
| // printf("queryDescriptors: cols %d\n" , queryDescriptors.cols); | ||
| // printf("queryDescriptors: type %d\n", queryDescriptors.type()); | ||
| // printf("queryPoints: %d\n", queryPoints.size()); | ||
|
|
||
| cv::Mat img_dst(dst._img->height, dst._img->width, CV_8UC3, dst._img->data); | ||
| cv::cvtColor(img_dst, img_dst, cv::COLOR_RGB2GRAY); | ||
| this->detector->detectAndCompute(img_dst, cv::Mat(), trainPoints, trainDescriptors); | ||
|
|
||
| // printf("trainDescriptors: rows %d\n", trainDescriptors.rows); | ||
| // printf("trainDescriptors: cols %d\n" , trainDescriptors.cols); | ||
| // printf("trainDescriptors: type %d\n", trainDescriptors.type()); | ||
| // printf("trainPoints: %d\n", trainPoints.size()); | ||
|
|
||
| if (trainPoints.size() == 0) | ||
| return ret; | ||
|
|
||
| std::vector<cv::DMatch> matches; | ||
| cv::BFMatcher matcher(cv::NORM_HAMMING, crossCheck); | ||
|
|
||
| //如果采用flannBased方法 那么 desp通过orb的到的类型不同需要先转换类型 | ||
| // cv::FlannBasedMatcher matcher(new cv::flann::LshIndexParams(table_number, key_size, multi_probe_level)); | ||
| // if (queryDescriptors.type() != CV_32F || trainDescriptors.type() != CV_32F) | ||
| // { | ||
| // queryDescriptors.convertTo(queryDescriptors, CV_32F); | ||
| // trainDescriptors.convertTo(trainDescriptors, CV_32F); | ||
| // } | ||
|
|
||
| matcher.match(queryDescriptors, trainDescriptors, matches); | ||
|
|
||
| if (matches.size() == 0) | ||
| return ret; | ||
|
|
||
| if (dump) // 避免之后 matches 顺序改变 | ||
| { | ||
| if (queryPoints.size() > 0) | ||
| { | ||
| cv::Mat imageOutput; | ||
| cv::drawKeypoints(img_src, queryPoints, imageOutput, cv::Scalar::all(-1), cv::DrawMatchesFlags::DRAW_RICH_KEYPOINTS); | ||
| cv::imwrite("cv_surf_img_src.jpg", imageOutput); | ||
| } | ||
| if (trainPoints.size() > 0) | ||
| { | ||
| cv::Mat imageOutput; | ||
| cv::drawKeypoints(img_dst, trainPoints, imageOutput, cv::Scalar::all(-1), cv::DrawMatchesFlags::DRAW_RICH_KEYPOINTS); | ||
| cv::imwrite("cv_surf_img_dst.jpg", imageOutput); | ||
| } | ||
| if (trainPoints.size() > queryPoints.size()) | ||
| { | ||
| cv::Mat imageOutput; | ||
| drawMatches(img_src, queryPoints, img_dst, trainPoints, matches, imageOutput); | ||
| cv::imwrite("cv_surf_img_matches.jpg", imageOutput); | ||
| } | ||
| } | ||
|
|
||
| //过滤出特征点距离 distance 小于 limit 以内的特征点 | ||
| for (int i = 0; i < matches.size(); i++) | ||
| { | ||
| // printf("distance: %f\n", matches[i].distance); | ||
| if (matches[i].distance < max) | ||
| { | ||
| max = matches[i].distance; | ||
| if (max < min) | ||
| max = min; | ||
| auto src_points = queryPoints[matches[i].queryIdx], dst_points = trainPoints[matches[i].trainIdx]; | ||
| auto tmp = py::make_tuple(matches[i].distance, (int)src_points.pt.x, (int)src_points.pt.y, (int)dst_points.pt.x, (int)dst_points.pt.y); | ||
| ret.append(tmp); | ||
| } | ||
| if (ret.size() >= limit) | ||
| break; | ||
| } | ||
| return ret; | ||
| } | ||
| }; | ||
|
|
||
| #endif |
This file was deleted.
Oops, something went wrong.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,5 +1,5 @@ | ||
|
|
||
| version='0.5.0' | ||
| version='0.5.1' | ||
|
|
||
| __all__ = ['display', 'camera', 'image'] | ||
|
|
||
|
|
||