https://github.com/jmlich/opencv-hog
Simple detector based on histogram of oriented gradients and support vector machines
https://github.com/jmlich/opencv-hog
computer-vision hog opencv svm
Last synced: about 1 month ago
JSON representation
Simple detector based on histogram of oriented gradients and support vector machines
- Host: GitHub
- URL: https://github.com/jmlich/opencv-hog
- Owner: jmlich
- Created: 2021-09-20T11:28:26.000Z (over 4 years ago)
- Default Branch: master
- Last Pushed: 2021-09-20T11:28:42.000Z (over 4 years ago)
- Last Synced: 2025-03-17T16:06:06.636Z (about 1 year ago)
- Topics: computer-vision, hog, opencv, svm
- Language: Python
- Homepage:
- Size: 837 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# opencv-hog
- extracts feature vector from image (see feature_vector function)
- "simple" hog
- hog_cv
- color histogram
- support vector machines classification
-- training set (first 80% of data) generates model
-- testing set (last 20% of data) evaluates model
- list of true/false positive/negative of testing set is printed in html to stdout
The input is defined as set of pictures with same resolution (e.g. 48x48) and class name
annotation file looks like this:
```
/var/www/html/fire/sun-dataset/SUNOUT/a/art_school/sun_aprywzqxjcezkbch.jpg,non-fire
/var/www/html/fire/labelme_samples-80/000093_65.jpg,fire
```
