Imaging Processing and Analysis with Applications(IPAA) I & II

IPAA: This team merges two previous VIP teams --Image Processing and Analysis (IPA) and Image-Based Mobile Phone Applications (APPS) --to balance the number of students in each section. Students can choose either Section I or II and work on similar projects. This team is eligible for senior design.

Description:

This team merges two previous VIP teams --Image Processing and Analysis (IPA) and Image-Based Mobile Phone Applications (APPS) --to balance the number of students in each section. Students can choose either Section I or II and work on similar projects. This team is eligible for senior design.

The team works on image processing and computer vision applications, with several projects that are in collaborations with faculties at different Schools and Departments at Purdue. These projects can also integrate mobile phone applications, and web-based GUI development to support the capture and visualization of visual information. Example projects include:

Mobile Phone Application Projects:

  • Style transfer using generative neural networks: Build an Android app that takes two input images and transfers the style of one image onto the other, generating a visually transformed output.
  • Traffic sign recognition: Develop an Android app that captures images containing traffic signs and recognizes them.
  • Synthetic image generation and detection: Create an Android app capable of generating synthetic images and identifying perturbations in each input image.
  • Panorama image generation from a collection of images: Design an Android app that can generate panorama images from uploaded files or live captures.

 

Image Processing and Analysis projects:

  • Continual learning for image classification: Develop a continual learning model for image classification using application-specific datasets to enable machine learning models to learn new information without forgetting previously learned knowledge, similar to how humans learn new things without forgetting past knowledge.
  • 3D food reconstruction from 2D images: Learn to use a 3D scanner and smartphone cameras to collect and process food objects and generate 3D views from 2D images.
  • Bone classification from custom datasets: Classify different types of bones from CT scans.
  • Automatically detect pedestrians: Classify the temporal behavior changes of pedestrians captured by surveillance cameras.

 

Graphic User Interface (GUI) Project:

  • Surgical Scene Segmentation: Develop a GUI-based platform for computational interpretation of surgical video data, addressing the challenge of annotating diverse anatomical structures by integrating semi-automated annotation and model training.

 

Relevant Texnologies:

  • Image Processing
  • Computer vision
  • Machine Learning
  • Mobile Phone Applications
  • Graphic User Interface design

 

Prerequisites:

  • Willingness to learn
  • Experience in the following topics is a plus:
    • Python programming, Android Programming, Git, machine learning/deep learning for computer vision, Graphic User Interface design, knowledge of statistics and visualization