Difference between revisions of "Image Sensor Interface and Pre-processing"
From iis-projects
(→Short Description) |
|||
| Line 1: | Line 1: | ||
[[File:Image Sensor Interface and Pre-processing.jpg|500px|thumb]] | [[File:Image Sensor Interface and Pre-processing.jpg|500px|thumb]] | ||
==Short Description== | ==Short Description== | ||
| − | In this project we will design a custom interface for | + | There is a continuous demand for high-speed image processing and recognition. Traditional machine vision systems are composed of image sensors and general-purpose processor.In this project we will design a custom interface for one or more image sensors. The interface will have a control unit that will allow the host system to control the parameters (image format, resolution, acquisition rate etc) of the image sensors. A data acquisition block will be used to access data from the image sensors. The main part of the project will be in designing the pre-processing stage which will perform a set of filtering and/or machine learning algorithms on the fly as the image is being acquired and forward this information to a larger host system for further processing. |
The project will require you to understand and implement the communication interface of common image sensors, will allow you to implement real-time video processing and machine learning algorithm steps directly ion hardware and interface with the host computer system. | The project will require you to understand and implement the communication interface of common image sensors, will allow you to implement real-time video processing and machine learning algorithm steps directly ion hardware and interface with the host computer system. | ||
| Line 31: | Line 31: | ||
: [http://www.iis.ee.ethz.ch/portrait/staff/lbenini.en.html Luca Benini] | : [http://www.iis.ee.ethz.ch/portrait/staff/lbenini.en.html Luca Benini] | ||
[[#top|↑ top]] | [[#top|↑ top]] | ||
| + | |||
==Detailed Task Description== | ==Detailed Task Description== | ||
Revision as of 15:55, 27 March 2014
Contents
Short Description
There is a continuous demand for high-speed image processing and recognition. Traditional machine vision systems are composed of image sensors and general-purpose processor.In this project we will design a custom interface for one or more image sensors. The interface will have a control unit that will allow the host system to control the parameters (image format, resolution, acquisition rate etc) of the image sensors. A data acquisition block will be used to access data from the image sensors. The main part of the project will be in designing the pre-processing stage which will perform a set of filtering and/or machine learning algorithms on the fly as the image is being acquired and forward this information to a larger host system for further processing.
The project will require you to understand and implement the communication interface of common image sensors, will allow you to implement real-time video processing and machine learning algorithm steps directly ion hardware and interface with the host computer system.
Status: Available
- Looking for 1-2 Semester/Master students
- Contact: Michele Magno
Prerequisites
- VLSI I
- VLSI II (recommended)
- Interest in video processing and machine learning
Character
- 20% System Design
- 60% ASIC Design
- 20% Verification
