Student Project Ideas
Augmented Reality in The Hounsfield FacilityThe University's Hounsfield Facility is a multidisciplinary research centre which uses novel imaging and image analysis methods to study the interactions between plant roots and the surrounding soil. Plant are grown in large soil columns in The Hounsfield glasshouse, imaged using a custom-built X-ray scanner and the resulting images segmented to separate plants from soil and produce a 3D model of the plant's (invisible) root system. The goal of this project is to develop an augmented reality tool, using the Vuforia package now found in Unity, to allow Hounsfield staff and visitors to the glasshouse to visualise the 3D root structure associated with a given plant. The first stage development will be a smartphone app, if successful the method will be demonstrated using a Microsoft HoloLens.
Learning to Position a Camera
Much of the work done in the Computer Vision Lab is concerned with recovering information about plants from colour images. The motivation for this is that the information we gather can be used along with genetic data to improve the productivity and resilience of crops to climate change, and ultimately provide more food. One difficulty is that plants are extremely complex objects and careful camera placement is vital. This project will use modern machine learning methods to create a system that is able to position a camera, held either by a 6 dog robot arm or a simpler x-y actuator table, in the position needed to allow reliable measurements to be made.
A Sensor Network for Environmental Monitoring
Environmental monitoring - of temperature, humidity, etc. - is important in a wide variety of contexts. Of particular interest at Nottingham is the monitoring and analysis of environmental conditions in research greenhouses. Our goal is the identification of new crops and varieties that can provide sustainable sources of food against the background of climate change; understanding of the conditions in which experiments are carried out is vital if this is to be achieved. The goal of this project is to build a network of wireless enabled environmental sensors and provide software that supports visualisation of the data they provide and the detection of outlier events that may affect the scientific results obtained.
Creating Super-Resolution Images
Images are one of the fastest growing forms of data, with millions being taken every day. Unfortunately, many of these are not of the resolution required by their owner, particularly those taken using older devices. Recovering a high-resolution image from a single low-resolution image is a classical problem in computer vision, and many approaches have been proposed. Recently, attention has turned to the use of deep machine learning methods such as Dong et al's SRCNN (https://arxiv.org/pdf/1501.00092.pdf). The goal of this project is to develop a software tool capable of creating super-resolution colour images, ideally using deep machine learning, and to evaluate that tool on a range of images.
What's that Chord?
There have been many attempts to analyse images of musical notation, e.g. to produce MIDI files that can be played or integrated into recordings. As a guitarist with limited music reading ability I have a very specific problem: I can read single notes at a reasonable speed but I find it hard to recognise chords when they are presented in standard notation. The goal of this project is to produce a smartphone app that can capture an image of a single chord expressed in standard notation (and maybe the key signature to provide context) and recognise which chord it is. Having recognised the chord, the app might play it, show alternative fingering diagrams, etc.
Image Style Transfer: Painterly Rendering
Much computer graphics work is aimed at producing photo-realistic images of artificial objects. Painterly rendering takes a different approach and aims to produce images which appear to be have been painted, by a particular artist, or in a particular artistic style. This is done by taking an initial, real image (or set of images) and manipulating it to produce a new image that appears painted. This project involves selecting an artist or style that interests you, and developing a software tool capable of rendering input images in that style.
3D Shape from Image Pairs
Binocular stereo - the recovery of 3D models from pairs of images - is a longstanding problem in computer vision. The Middlebury Stereo site (http://vision.middlebury.edu/stereo/) provides several sets of image pairs, along with ground truth data showing what the true 3D structure of the scene is. This is a valuable resource for those attempting to develop new stereo methods. The goal of this project is to atempt the Middlebury challenge - to develop software capable of extracting some 3D measurements from a stereo image pair, and compare it to the ground truth.
Input image Ground truth depth map