patternfinders-141205004957-conversion-gate02.pptx

cifar 10

1. CIFAR- 10 Object Recognition in Images

2. Background Image Classification Applications Automatic image annotation Reverse image search Kinds of datasets Digital Images Few thousands to millions of images. Machine Learning problems in this domain Image search engine Face detection

3. Problem Task is to classify images into following ten labels. The data labels are Cat, Dog, Airplane, Automobile, Bird, Deer, Frog, Horse, Ship, Truck. Challenges Large Data set Small Image size (32x32)

4. Dataset Data consists of 60,000 32x32 color images in 10 classes, with 6000 images per class. There are 50,000 training images and 10,000 test images The classes are completely mutually exclusive. With the above mentioned samples, we will build a model to identify whether the given image is Cat or a Dog etc.

5. Data Insights The objects within classes in this dataset are extremely varied. For example the "bird" class contains many different types of bird (both big birds and small). Not only are there many types of bird, but the occur at many possible magnifications, all possible angles and all possible poses. Sometimes only parts of the bird are shown. The poses problem is even worse for the dog/cat category, because these animals occur at many different types of poses, and sometimes only the head is shown or left part of the body, etc.

6. Method 1 Divide every 32*32 px image into 64 patches of 4*4 px Feature length would be 3*16*64 = 3072. Followed by Fisher on all images and extract their top 9 Dimensions Values obtained above will represent final feature vector for images. Classification using SVM/kNN.

7. Method 1 (Continue…) 4 4 32 32 FDA Patches SVM/kNN Patches FDA SVM /kNN Model Labels Block diagram Train Data Test Data

8. Method 1 Drawbacks This method based on just intensity. Same object captured in different light. Image taken from different angles.

9. Method 2 Aim : Preserve the Shape of Image Step 1 : Image is filtered using Laplace of Gaussian filter(LoG). log_image = edge(gray_image,'log', [], 2);

10. Method 2 (Continue…) Step 2 : Feature is constructed by using histogram of oriented gradients gives 324- d feature vector hog_image= hog_feature_vector(log_image)

11. Method 2 (Continue…) FDA LoG and HoG SVM/kNN LoG and HoG FDA SVM/kNN Model Labels Block Diagram Train Data Test Data

12. Experiments Divide given 50K data to train data and test data. Apply Feature engineering and classification. 40,000(80%) images as train data, 10,000(20%) images as validation data. Result Evaluation : Percentage of labels that are predicted correctly Number of Fisher dimensions (D = 7, 9, 13) SVM kernel (linear, rbf) kNN( k= 13, 17, 21)

13. Results Benchmark : If we label all image as cat, accuracy is 10% Classifier # Fisher Dimentions Method 1 Method 2 KNN (k=21) 7 20.59 25.65 KNN (k=21) 9 22.43 28.34 KNN (k=21) 13 22.68 28.71 SVM 7 30.98 40.45 SVM 9 34.56 42.85 SVM 13 34.83 -

14. Thank You