DSPRelated.com
Forums

image compression

Started by natasha adams January 6, 2006
hello all:

I'm currently looking for any error concealment code for image transmissions
that will contain redundancy which will enable me to vary its amount so that
i can compare redundacy with amount of error correction that has to be added
into the code.

Also if anyone is aware of an image compression algorithm and its soure
code, please do let me know. I have written the code for J2K but i'm looking
for something simpler. Any help is greatly appreciated. Thanks in advance,
Regards,
natasha



Hi , The diff. may be as followa(Though it is not in tech. terms) This specification describes an image compression system that allows great flexibility, not only for the compression of images, but also for the access into the compressed data. The codestream provides a number of mechanisms for locating and extracting data for the purpose of retransmission, storage, display, or editing. This access allows storage and retrieval of data appropriate for a given application, without decoding. > >I m working on image compression for my final training course. I am >asked to compress the images 1600*1200 the most I can with the less >loss I can... > > I am working with matlab, and trying to understand jpeg and jpeg2000. > >In fact my boss wants me to use jpeg2000 but EBCOT is very difficult. >So i wanted to know, what makes jpeg2000 so different from jepg except >DWT ? And what can I add in jpeg algo to improve the results, thats to >say, the ratio and the loss ? (I don't want to use jpeg with no loss). >I know it s not a very technical question but I hope sbdy will have >answers for me. > >Thanx