https://github.com/defi0x1/Classification-SIFT-SVM
Build a classifier to classify transport using sift and svm
https://github.com/defi0x1/Classification-SIFT-SVM
bag-of-words classification gridsearchcv sift-algorithm sift-descriptors svm svm-classifier
Last synced: about 1 month ago
JSON representation
Build a classifier to classify transport using sift and svm
- Host: GitHub
- URL: https://github.com/defi0x1/Classification-SIFT-SVM
- Owner: defi0x1
- Created: 2019-06-21T11:17:14.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2021-12-26T06:42:08.000Z (almost 4 years ago)
- Last Synced: 2025-07-21T14:50:54.983Z (3 months ago)
- Topics: bag-of-words, classification, gridsearchcv, sift-algorithm, sift-descriptors, svm, svm-classifier
- Language: Python
- Homepage:
- Size: 31.1 MB
- Stars: 10
- Watchers: 0
- Forks: 3
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Classification-SIFT-BoW-SVM
Build a classifier to classsification transport using sift, bag of words and svm# Tree Project
Root/
data # Data contains total 1854 image of 5 classes: bus, car, moto, pedestrian
data_loader.py # Imread image and label for data
sift_extractors.py # Extract features with SIFT and Build Bag of Word
build_model.py # Build model SVM and training
gridSearchCV # implement GridSearchCV find Hyper-params
# Hyperparameters
- Number of centroid in Kmean: 60
- Penalty parameter C of the error term in SVM: 30# Label
- 'moto': 0
- 'car': 1
- 'pedestrian': 2
- 'truck': 3
- 'bus': 4
# Statistic Data
- Statistics All Classes
![]()
- Statistics Classes
![]()
# Accuracy
- Accuracy score: 72.9% - test size: 0.2
# Using GridSearchCV
- Best params
'clf__C': 5,
'clf__gamma': 0.01,
'clf__kernel': 'rbf'
- Accuracy: 75.72% +/-1.52%
# Requirements
- opencv-contrib-python==3.4.2.16
- opencv-python==3.4.2.16