Objectives of Project: Solving a real-world problem by designing, implementation, and comparative analysis of algorithms in terms of accuracy, time and memory complexities.

Problem Statement: Searching image in a large dataset effectively and efficiently is the aim of this project. The image search should be based on image content not label or file name. We want to design an algorithm, which generates a barcode for all images in the dataset and when we search an image, we compare the barcode of the current under query image with the barcode of all other images in the dataset to find best matching image. Our both algorithms, which generate the barcode and conduct the search based on the barcode, should be fast. Our designed barcode should be short in size. Our utilized dataset is the MNIST, 70,000 handwritten digits images. The deigned algorithms should be applicable on other datasets with variant image types and dataset sizes, such as millions of x-ray images captured from different parts of human body. Required Algorithms: Our two main algorithms are Barcode_Generator algorithm to generate barcode for an image and Search_Algorithm to search a query image in the dataset. You need to compare your algorithm with Radon based barcode method. You will try to design algorithms which generate short barcode and offers fast, and accurate searches.

Required Measurements and Analysis: You need to compare your method and Radon based method on the following items: 1)Time in seconds to generate barcode for all 70,000 images in MINST dataset, 2)Time in seconds to search all 70,000 images to find the best matching image for each image, 3)Accuracy in term of hit ratio (over all 70,000 image search what percentage of search is correct; i.e., retrieved image is from the same class of the query images), 4)Big-oh complexity analysis for Barcode_Generator and Search_Algorithm.

Deliverables: A comprehensive report with the source codes should be submitted. The report should cover: A) Algorithms Explanation in detail, B) Required Measurements and Analysis, C)Results Comparison, D)Search Results (at least 3 search per class), and E) Conclusion Remarks.

Project Delivery: The project will be delivered to the course TA. The TA will ask some questions from all group members. All group members should be involved in all components of project.

Project Presentation: Project presentations will be done in the class. Dates will be announced in the CANVAS.

Note: You need to compare your designed algorithm with the Radon based algorithm. Your designed barcode can be generated in any method which you prefer, there is no a limitation.

PLEASE DONT COMPLETE AND DONT FEEL LIKE YOU HAVE TO COMPLETE BUT I GENUINELY DONT KNOW WHERE TO START AND DESPRETLY NEED HELP LITERALLY JUST TO START THIS PROJECT! WE HAVE NOT LEARNED ENOUGH TO DO THIS AND I AM STRESSING