| Author |
Message |
cedance
Joined: 24 Oct 2003
Posts: 704
Helped: 28
Location: Germany
|
 10 Oct 2004 7:09 image processing |
|
|
|
hi,
could someone spot me out the challenges and the fields that are open with regard to image processing.. i have already PM ed some of my friends in this forum reg, this. however i thought askin the board will lead to many responses which will also prove useful to others....
regards,
arunmit168.
|
|
| Back to top |
|
 |
nicleo
Joined: 06 Sep 2004
Posts: 717
Helped: 60
|
| 10 Oct 2004 11:49 image processing |
|
|
|
| AI-Content Based Image Retrieval
|
|
| Back to top |
|
|
cedance
Joined: 24 Oct 2003
Posts: 704
Helped: 28
Location: Germany
|
| 10 Oct 2004 19:18 Re: image processing |
|
|
|
| nicleo wrote: |
| AI-Content Based Image Retrieval |
could u explain that definitive more detailed.. it would be useful for me as well as others... thank u...
regards,
arunmit168,
|
|
| Back to top |
|
|
djalli
Joined: 10 Nov 2001
Posts: 887
Helped: 15
Location: 1600 Pennsylvania Avenue, Washington DC 20500
|
| 10 Oct 2004 21:04 image processing |
|
|
|
There are thousands, there is big money in medical imagery.
Let suppose you do an MRI, you do not have precise sample data to compare but you need more neural imagery analysis to approximate a better educated answer.
Normally people who work here are not that very experienced and doctors will not have a great deal of understanding throughly what is indeed behind the newly collected sample data even if it looks similar to a previous sample.
In few decades microbes have become resistant to antibiotics while few decades ago were very effective. It is a learning process even with microbes and viruses. A great deal of analysis in imagery is indeed a better fine grained image processing.
In the way you would find more and more problems as they are as of now thousands of challenges yet unsolved.
Aim at medical imagery processing.
|
|
| Back to top |
|
|
nicleo
Joined: 06 Sep 2004
Posts: 717
Helped: 60
|
| 11 Oct 2004 3:51 Re: image processing |
|
|
|
| arunmit168 wrote: |
| could u explain that definitive more detailed.. it would be useful for me as well as others... thank u... |
| Quoted from Content-Based Image Retrieval Systems - A Survey wrote: |
| In many areas of commerce, government, academia, and hospitals, large collections of digital images are being created. Many of these collections are the product of digitizing existing collections of analogue photographs, diagrams, drawings, paintings, and prints. Usually, the only way of searching these collections was by keyword indexing, or simply by browsing. Digital images databases however, open the way to content-based searching... more details, pls visit the links at below. |
References
Content-Based Image Retrieval Systems: A Survey
http://www.aa-lab.cs.uu.nl/cbirsurvey/cbir-survey/cbir-survey.html
Content-Based Image Retrieval for Medical Image Databases
http://rvl2.ecn.purdue.edu/~cbirdev/WWW/CBIRmain.html
Potential Applications:
Medical image processing
Satelite image processing
Crime prevention (e.g. airport security surveillance system)
Web searching
Intellectual property such as trademark registration
Architectural and engineering design
Fashion and interior design
Journalism and advertising
Artificial Intelligent algorithms (e.g. neural network and fuzzy logic) can be used to improve the performance and robustness of existing/future image retrieval systems.
You can also see "Text-Based Image Retrieval System", such as
Google Image Search
http://images.google.com/imghp?hl=en
Try the following search keyword: cancer
Search results: 342,000 hits (10 Oct 2004)
Besides cancer (related to diseases), the search results also contained cancer (crab, zodiac). So, if you're expected for cancer (diseases), then Google Image Search did not do the work well. If you can modify the system to include content-based feature (e.g. shape, color, etc), you will get better or more accurate results.
|
|
| Back to top |
|
|
videohu
Joined: 06 Oct 2004
Posts: 70
|
| 14 Oct 2004 21:22 image processing |
|
|
|
| Maybe, machine learning, artificial intelligence will play great role in these area, especially in pattern classification. If you want to pursue them very deeply, you should have strong foundation of mathematic such as statistical theory, martrix theory, convex optimization and function analyiss.Good luck.
|
|
| Back to top |
|
|
