LAB 6 REPORT - SAISRIHARSHAS/Big-Data-Analytics-and-Applications-CS5542 GitHub Wiki

Write a spark program for the following Machine Learning Task. Create your own dataset for Image Classification/Object Detection Problem. Handle fuzzy classification/object detection task using at least two classification algorithms (e.g., Decision Tree, Random Forest, Naïve Bayes). Report the accuracy and confusion matrix obtained.

IMPLEMENTATION: DATA INPUT: https://github.com/SAISRIHARSHAS/Big-Data-Analytics-and-Applications-CS5542/tree/master/Lab%206/Source/SparkProgramming/FeatureExtraction

The dataset is a video about wild animals. Used SIFT to get detailed accuracy of the objects in the video. Used fuzzy classification to classify the images. The red line around the objects indicates that the given image is being recognized in the video.

Object Detection:

Obtained accuracy along with final confusion matrix using both Random Forest and Decision Tree model.

Decision Tree (accuracy and confusion matrix):

Random Forest (accuracy and confusion matrix):

Sample Code:

Write an Android application 1)Image classification (fuzzy) and object detection through the Spark API 2)Image classification/object detection using Clarifei API https://www.clarifai.com/api Refer to Tutorial 6 for fuzzy image classification/object detection Refer to Tutorial 5 Spark API tutorial. Refer to Tutorial 3 Clarifei API tutorial.

1.Created a simple application to display the accuracy from both Random Forest and Decision Tree. Used SimpleHttpServer to host a local server and showed the accuracy in the application. Application Home Page:

Random Forest & Decision Tree Accuracy:

2. Created a simple application to annotate images. Used the mainframes from tutorial 3 Video Annotation in Android Application. Application: