Closed mqchau closed 10 years ago
mqchau
commented
10 years ago Hi guys, When you write your function such as IsolateCharacter(IMAGE *...), please put them in a separate file instead of stuffing everything into Model.c Last time Andrew had a lot of difficulty at last minute trying to merge Model.c. This time we will avoid that by splitting it into multiple small files to reduce chances of merge conflicts. Quan.
mqchau
commented
10 years ago Eric, I just thought of this. When you make your crop image function, you'll get (x1, y1) and (x2, y2). But please make it so that they could be arbitrary number inside the picture boundary please. This means point 1 could be either left or right, top or bottom of the point 2. Thank you. quan.
mqchau
commented
10 years ago How's everybody doing on your programming? I knew about Kevin and Ryan. What's your progress status now?
antrinh
commented
10 years ago I'm just doing a lot of studying. I just finished reading the academic article you sent me yesterday, and it gave me some good insight. I have a question though, in the academic article, the image is cropped and resized and whatnot for proper comparison with a template library. Now I was wondering, what exactly is the scope of my method? Is it only comparing and then sending a utarrray with character match stats to kevin for postprocessing?
On Sun, Feb 23, 2014 at 10:54 AM, mqchau notifications@github.com wrote:
How's everybody doing on your programming? I knew about Kevin and Ryan. What's your progress status now?
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35839366 .
mqchau
commented
10 years ago Is it only comparing and then sending a utarrray with character match stats to kevin for postprocessing? Agree Input: ILIST Output: CharProfile *
For starter, you can try to do this: Read a template char image in, add some noise to it. Method AddNoise could be taken from last quarter. Then try to identify the new image. Quan.
On Sun, Feb 23, 2014 at 11:52 AM, Andrew Trinh notifications@github.comwrote:
I'm just doing a lot of studying. I just finished reading the academic article you sent me yesterday, and it gave me some good insight. I have a question though, in the academic article, the image is cropped and resized and whatnot for proper comparison with a template library. Now I was wondering, what exactly is the scope of my method? Is it only comparing and then sending a utarrray with character match stats to kevin for postprocessing?
On Sun, Feb 23, 2014 at 10:54 AM, mqchau notifications@github.com wrote:
How's everybody doing on your programming? I knew about Kevin and Ryan. What's your progress status now?
Reply to this email directly or view it on GitHub< https://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35839366> .
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35841118 .
mqchau
commented
10 years ago Sorry, my mistake. Input: ILIST Output: UT_array *, this array is a bunch of CharProfile
On Sun, Feb 23, 2014 at 12:01 PM, Quan Quoc Minh Chau mqchau@uci.eduwrote:
Is it only comparing and then sending a utarrray with character match stats to kevin for postprocessing? Agree Input: ILIST Output: CharProfile *
For starter, you can try to do this: Read a template char image in, add some noise to it. Method AddNoise could be taken from last quarter. Then try to identify the new image. Quan.
On Sun, Feb 23, 2014 at 11:52 AM, Andrew Trinh notifications@github.comwrote:
I'm just doing a lot of studying. I just finished reading the academic article you sent me yesterday, and it gave me some good insight. I have a question though, in the academic article, the image is cropped and resized and whatnot for proper comparison with a template library. Now I was wondering, what exactly is the scope of my method? Is it only comparing and then sending a utarrray with character match stats to kevin for postprocessing?
On Sun, Feb 23, 2014 at 10:54 AM, mqchau notifications@github.com wrote:
How's everybody doing on your programming? I knew about Kevin and Ryan. What's your progress status now?
Reply to this email directly or view it on GitHub< https://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35839366
.
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35841118 .
ghost
commented
10 years ago I've finished and tested rotate, and really just used last quarter's functions for crop and resize (btw, do I need to worry about aspect ratio?). Color filter I think I've finished, just need to test it.
-Eric
On Feb 23, 2014, at 10:54 AM, mqchau notifications@github.com wrote:
How's everybody doing on your programming? I knew about Kevin and Ryan. What's your progress status now?
— Reply to this email directly or view it on GitHub.
mqchau
commented
10 years ago For cropping, could you make sure that it will work when first coordinate is in left or right, above or below the second coordinate?
On Sun, Feb 23, 2014 at 12:08 PM, ericr2 notifications@github.com wrote:
I've finished and tested rotate, and really just used last quarter's functions for crop and resize (btw, do I need to worry about aspect ratio?). Color filter I think I've finished, just need to test it.
-Eric
On Feb 23, 2014, at 10:54 AM, mqchau notifications@github.com wrote:
How's everybody doing on your programming? I knew about Kevin and Ryan. What's your progress status now?
Reply to this email directly or view it on GitHub.
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35841645 .
ghost
commented
10 years ago Yea, I thought that was implied. Pretty sure I already took care of it.
-Eric
On Feb 23, 2014, at 1:43 PM, mqchau notifications@github.com wrote:
For cropping, could you make sure that it will work when first coordinate is in left or right, above or below the second coordinate?
On Sun, Feb 23, 2014 at 12:08 PM, ericr2 notifications@github.com wrote:
I've finished and tested rotate, and really just used last quarter's functions for crop and resize (btw, do I need to worry about aspect ratio?). Color filter I think I've finished, just need to test it.
-Eric
On Feb 23, 2014, at 10:54 AM, mqchau notifications@github.com wrote:
How's everybody doing on your programming? I knew about Kevin and Ryan. What's your progress status now?
Reply to this email directly or view it on GitHub.
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35841645 .
— Reply to this email directly or view it on GitHub.
mqchau
commented
10 years ago Good. To commit your changes, you need to the following step:
Quan
On Sun, Feb 23, 2014 at 2:28 PM, ericr2 notifications@github.com wrote:
Yea, I thought that was implied. Pretty sure I already took care of it.
-Eric
On Feb 23, 2014, at 1:43 PM, mqchau notifications@github.com wrote:
For cropping, could you make sure that it will work when first coordinate is in left or right, above or below the second coordinate?
On Sun, Feb 23, 2014 at 12:08 PM, ericr2 notifications@github.com wrote:
I've finished and tested rotate, and really just used last quarter's functions for crop and resize (btw, do I need to worry about aspect ratio?). Color filter I think I've finished, just need to test it.
-Eric
On Feb 23, 2014, at 10:54 AM, mqchau notifications@github.com wrote:
How's everybody doing on your programming? I knew about Kevin and Ryan. What's your progress status now?
Reply to this email directly or view it on GitHub.
Reply to this email directly or view it on GitHub< https://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35841645>
.
Reply to this email directly or view it on GitHub.
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35846151 .
hanchelc
commented
10 years ago You asked for the status of my code, and I will tell you that nothing has been done on it. HOWEVER, I am a reliable human being and I will have it done before the day ends tomorrow! Is that too late?
mqchau
commented
10 years ago Not too late. I trust your reliability.
On Sun, Feb 23, 2014 at 3:53 PM, hanchelc notifications@github.com wrote:
You asked for the status of my code, and I will tell you that nothing has been done on it. HOWEVER, I am a reliable human being and I will have it done before the day ends tomorrow! Is that too late?
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35848606 .
hanchelc
commented
10 years ago Update to the status of my code: I have a draft written, but no test has been done on it yet nor has the code been compiled.. I just wrote it on paper. How can I get an IMAGE * to test with? ^^;
mqchau
commented
10 years ago ReadImage could be taken from FileIO.c of last quarter. https://eee.uci.edu/13f/18040/assignment.html You can read EECS22L_Project2 https://github.com/mqchau/EECS22L_Project2 / Imageshttps://github.com/mqchau/EECS22L_Project2/tree/master/Images / _CourierNew12300DPI.ppm as a example.
On Mon, Feb 24, 2014 at 3:16 PM, hanchelc notifications@github.com wrote:
Update to the status of my code: I have a draft written, but no test has been done on it yet nor has the code been compiled.. I just wrote it on paper. How can I get an IMAGE * to test with? ^^;
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35952536 .
hanchelc
commented
10 years ago Ahhh! Thanks!
duongkt
commented
10 years ago wow what the hell....
"Good Job! The plans for different phase and branchings are well described. No testing plans. Please add your used algorithm for cropping and similarity check to final report."
yet we got a 85/100!?!?!?!?
On Mon, Feb 24, 2014 at 3:30 PM, hanchelc notifications@github.com wrote:
Ahhh! Thanks!
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35953781 .
rmorrison23
commented
10 years ago It's the testing (or lack thereof)—they criticized us last time for not having it, and in this project they listed it specifically as a requirement. I'm assuming this is something that all teams are deficient in.
duongkt
commented
10 years ago I thought we do have test plan kinda... Everyone is responsible for testing their part separately before uploading to git master branch since we are working on our own individual part before combining it. Then we can test it with actual files and stuff instead of making our own.
On Mon, Feb 24, 2014 at 5:37 PM, rmorrison23 notifications@github.comwrote:
It's the testing (or lack thereof)--they criticized us last time for not having it, and in this project they listed it specifically as a requirement. I'm assuming this is something that all teams are deficient in.
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35965409 .
mqchau
commented
10 years ago That branching technique is something I learned from a textbook at work. Advance stuff. I can explain it to you guys if you guys want.
On Mon, Feb 24, 2014 at 5:59 PM, duongkt notifications@github.com wrote:
I thought we do have test plan kinda... Everyone is responsible for testing their part separately before uploading to git master branch since we are working on our own individual part before combining it. Then we can test it with actual files and stuff instead of making our own.
On Mon, Feb 24, 2014 at 5:37 PM, rmorrison23 <notifications@github.com
wrote:
It's the testing (or lack thereof)--they criticized us last time for not
having it, and in this project they listed it specifically as a requirement. I'm assuming this is something that all teams are deficient in.
Reply to this email directly or view it on GitHub< https://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35965409>
.
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35966743 .
hanchelc
commented
10 years ago Quan... how do I get the Project2 repository to show up in my directories...?
mqchau
commented
10 years ago git clone https://github.com/hanchelc/EECS22L_Project2.git
On Mon, Feb 24, 2014 at 10:13 PM, hanchelc notifications@github.com wrote:
Quan... how do I get the Project2 repository to show up in my directories...?
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35978344 .
hanchelc
commented
10 years ago I tried that 10 minutes ago!
Something about a fatal repository and not being able to find it.
mqchau
commented
10 years ago what does it say?
On Mon, Feb 24, 2014 at 10:20 PM, hanchelc notifications@github.com wrote:
I tried that 10 minutes ago!
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35978626 .
hanchelc
commented
10 years ago Wait. Should I be logged out of Group1 user?....
mqchau
commented
10 years ago Try this git clone git@github.com:hanchelc/EECS22L_Project2.git
On Mon, Feb 24, 2014 at 10:22 PM, Quan Quoc Minh Chau mqchau@uci.eduwrote:
what does it say?
On Mon, Feb 24, 2014 at 10:20 PM, hanchelc notifications@github.comwrote:
I tried that 10 minutes ago!
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35978626 .
hanchelc
commented
10 years ago "ERROR: Repository not found. fatal: Could not read from remote repository.
Please make sure you have the correct access rights and the repository exists."
mqchau
commented
10 years ago I'm trying to chat with your account trufflelove@gmail.com Hanchel
On Mon, Feb 24, 2014 at 10:23 PM, hanchelc notifications@github.com wrote:
Wait. Should I be logged out of Group1 user?....
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35978739 .
duongkt
commented
10 years ago hey hanchel i had that problem too. Move to the directory that is just 1 up from the eecs22_project1 then try the git clone again. I went too far up once and i get that error.
On Mon, Feb 24, 2014 at 10:26 PM, mqchau notifications@github.com wrote:
I'm trying to chat with your account trufflelove@gmail.com Hanchel
On Mon, Feb 24, 2014 at 10:23 PM, hanchelc notifications@github.com wrote:
Wait. Should I be logged out of Group1 user?....
Reply to this email directly or view it on GitHub< https://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35978739>
.
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35978909 .
hanchelc
commented
10 years ago The first thing I tried was to git clone in the EECS22L directory. ):
duongkt
commented
10 years ago weird... It was fine when i made a new folder and use git clone while in that directory on the command line.
On Mon, Feb 24, 2014 at 10:29 PM, hanchelc notifications@github.com wrote:
The first thing I tried was to git clone in the EECS22L directory. ):
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-35979060 .
antrinh
commented
10 years ago Hello teammates, I will not be attending the lab session today because I need to catch up on some other work. I will work on the project in the Thursday lab session though.
mqchau
commented
10 years ago I'm combining Alpha Release and here is the status of parts:
antrinh
commented
10 years ago Saturday night at the latest
On Fri, Feb 28, 2014 at 11:41 PM, mqchau notifications@github.com wrote:
I'm combining Alpha Release and here is the status of parts:
- ReadImage: pending from Jamie
- Preprocessing: got crop + rotate: OK to merge
- Isolate character: pending from Hanchel
- Match character: pending from Andrew
- Postprocessing: taking the char with highest probability: OK to merge
- GUI: layout working: OK to merge Jamie, Hanchel and Andrew please submit your part asap so I could continue with merging. Quan.
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-36418830 .
hanchelc
commented
10 years ago Teeeeeam. I pushed what I think is a pretty good Isolate Character for alpha release. To see what it spits out, switch to the IsolateCharacters branch, change directory to ~/EECS22/EECS22L_Project2/src/Model/, make clean, make TestIsolate, ./TestIsolateCharacters, and then hundreds of test*.ppm images should appear in the current folder. The test0.ppm should be 'A,' and then the next should be 'B,' etc. There are a number of blanks because my code will start from the first left-most character it detects, in this case 'A,' and just crop squares of char_height (determined by font size and dpi) and char_width, determined by measuring distances between the white space in between characters of the monospaced Courier New font. Ignoring the whitespaces, most of the letters do look good. I am still having an issue differentiating q and g since they get their upper half cropped and their tails get left behind. Soooo, let me know if you have more .ppms to test with and I'll take a shot at them with my code. (: Otherwise, you can feel free to discuss any comments or concerns you have with my current code.
antrinh
commented
10 years ago Hello all,
I was wondering if anyone knows whether or not there is already animagelist containing all of the images corresponding to the CourierNew12_300DPI file somewhere that I could use. If there isn't how would I go about populating that list? I need this so I can make the comparisons.
-Andrew
On Sat, Mar 1, 2014 at 8:38 AM, hanchelc notifications@github.com wrote:
Teeeeeam. I pushed what I think is a pretty good Isolate Character for alpha release. To see what it spits out, switch to the IsolateCharacters branch, change directory to ~/EECS22/EECS22L_Project2/src/Model/, make clean, make TestIsolate, ./TestIsolateCharacters, and then hundreds of test*.ppm images should appear in the current folder. The test0.ppm should be 'A,' and then the next should be 'B,' etc. There are a number of blanks because my code will start from the first left-most character it detects, in this case 'A,' and just crop squares of char_height (determined by font size and dpi) and char_width, determined by measuring distances between the white space in between characters of the monospaced Courier New font. Ignoring the whitespaces, most of the letters do look good. I am still having an issue differentiating q and g since they get their upper half cropped and their tails get left behind. Soooo, let me know if you have more .ppms to test with and I'll take a ! shot at t hem with my code. (: Otherwise, you can feel free to discuss any comments or concerns you have with my current code.
You may or may not have to include
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-36429239 .
hanchelc
commented
10 years ago Run my test like I explained above. You should get hundreds of .ppms to test with (most of which will be blank though. Just save the ones that have characters cropped onto your desktop). There's also one imagelist that my test produces that you can grab.
antrinh
commented
10 years ago Ah ok. Also, we are missing an open quotation mark while we have a closing one.
On Sat, Mar 1, 2014 at 3:24 PM, hanchelc notifications@github.com wrote:
Run my test like I explained above. You should get hundreds of .ppms to test with. There's also one imagelist that my test produces that you can grab.
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-36440313 .
antrinh
commented
10 years ago Also Hanchel, since the ppms are labeled test0.ppm, test1.ppm .... testn.ppm, how do I know which one corresponds to what character? Do I need to look at each ppm to see what image they are so that I can match them with an actual char?
On Sat, Mar 1, 2014 at 3:43 PM, Andrew Nam Trinh antrinh@uci.edu wrote:
Ah ok. Also, we are missing an open quotation mark while we have a closing one.
On Sat, Mar 1, 2014 at 3:24 PM, hanchelc notifications@github.com wrote:
Run my test like I explained above. You should get hundreds of .ppms to test with. There's also one imagelist that my test produces that you can grab.
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-36440313 .
hanchelc
commented
10 years ago Yes. :D It shouldn't be too bad. test0, 1, and 2 are definitely A, B, and C.
hanchelc
commented
10 years ago Is that too annoying? don't even take in each ppm at a time. Just take in my imglist. Can you just include that bit of my model.c that generates the imglist in your test code?
hanchelc
commented
10 years ago Andrew you said previously: "Ah ok. Also, we are missing an open quotation mark while we have a closing one." Is this an issue? I don't understand where or even what this problem is.
mqchau
commented
10 years ago Under fonts directory
On Saturday, March 1, 2014, Andrew Trinh notifications@github.com wrote:
Hello all,
I was wondering if anyone knows whether or not there is already an imagelist containing all of the images corresponding to the CourierNew12_300DPI file somewhere that I could use. If there isn't how would I go about populating that list? I need this so I can make the comparisons.
-Andrew
On Sat, Mar 1, 2014 at 8:38 AM, hanchelc notifications@github.com<javascript:_e(%7B%7D,'cvml','notifications@github.com');> wrote:
Teeeeeam. I pushed what I think is a pretty good Isolate Character for alpha release. To see what it spits out, switch to the IsolateCharacters branch, change directory to ~/EECS22/EECS22L_Project2/src/Model/, make clean, make TestIsolate, ./TestIsolateCharacters, and then hundreds of test*.ppm images should appear in the current folder. The test0.ppm should be 'A,' and then the next should be 'B,' etc. There are a number of blanks because my code will start from the first left-most character it detects, in this case 'A,' and just crop squares of char_height (determined by font size and dpi) and char_width, determined by measuring distances between the white space in between characters of the monospaced Courier New font. Ignoring the whitespaces, most of the letters do look good. I am still having an issue differentiating q and g since they get their upper half cropped and their tails get left behind. Soooo, let me know if you have more .ppms to test with and I'll take a ! shot at t hem with my code. (: Otherwise, you can feel free to discuss any comments or concerns you have with my current code.
You may or may not have to include
Reply to this email directly or view it on GitHub< https://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-36429239> .
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-36440202 .
hanchelc
commented
10 years ago Oh I see. A reference image of an open quote and a reference image of a close quote.
hanchelc
commented
10 years ago whoops my bad. QUAN. please answer the question I left in gchat.
antrinh
commented
10 years ago Sure, but the thing is, how do I know what order the symbols are in?
On Sat, Mar 1, 2014 at 4:07 PM, hanchelc notifications@github.com wrote:
Closed #3 https://github.com/mqchau/EECS22L_Project2/issues/3.
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3 .
hanchelc
commented
10 years ago I just realized when I commented out my debugging printfs, I might've missed the ' if ' statements before them. that'll fk up the code if you try to run it.
anyway, uhhh... how do you know what order? test0, test1, ..., testn will go in the order Left to Right, Up to Down, starting from A. Do you get what I mean?
mqchau
commented
10 years ago Andrew, you should just focus on determine only 1 image from Hanchel. Don't worry about order for now. Order will just be an upper level function iterating through every image, calling a lower level function comparing each images with the fonts database.
On Sat, Mar 1, 2014 at 4:14 PM, hanchelc notifications@github.com wrote:
I just realized when I commented out my debugging printfs, I might've missed the ' if ' statements before them. that'll fk up the code if you try to run it.
anyway, uhhh... how do you know what order? test0, test1, ..., testn will go in the order Left to Right, Up to Down, starting from A. Do you get what I mean?
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-36441436 .
jamiesl
commented
10 years ago Hey Quan,
I'm pretty sure my code is solid, but when I try to compile for testing, it says "undefined reference to [whatever function]" although I pasted Image.h into the inc files and Image.c into the src files. I made sure to include them in my header and also compiled using "gcc TestingReadImage.c -o Test -I../../inc/uthash -I../../inc/Model". Any ideas?
Thanks
jamiesl
commented
10 years ago HAH kidding. Just realized they're in separate files, so I have to link them... :v
mqchau
commented
10 years ago I was typing that answer to you.
On Sat, Mar 1, 2014 at 9:45 PM, jamiesl notifications@github.com wrote:
HAH kidding. Just realized they're in separate files, so I have to link them... :v
Reply to this email directly or view it on GitHubhttps://github.com/mqchau/EECS22L_Project2/issues/3#issuecomment-36446898 .
Hi all, Please find below the description of Alpha Release. Our goal here is to earn all basic credit for Alpha Release + small extra credit of supporting extra images format. You'll need to read so you don't develop in wrong direction. I know it's long but there's a lot of empty spaces.
https://drive.google.com/a/uci.edu/file/d/0B6hFDKbQXelUamtkdnFMZllhaHM/edit?usp=sharing
Quan.