Monday 12 May 2014

The 'Joy' of Double Rejection


Hell ya! Today's entry is about the 'joy' of double rejection. I bet many of you have been rejected before or probably I should say unsuccessful rather than that nasty word. There is no joy of being rejected of course. I my previous post I talked about how was my feeling after my ICPR paper has been rejected. In fact, I said I managed to get better results. Well, it was true. But the saddest part is as soon as I posted the entry, I received an email regarding to my MIUA paper decision. I must say it was the earliest paper decision I have every received in the last two years. In all conferences I have attended or submitted a paper before, all of them extended the 'Author Notification' date to one or two days late. Initially when I open my uni email, I got an email with a header something like 'Decision paper number 49' and I thought it can't be because it was about 3 days earlier than the date posted on the conference's website. Then I slowly opened it and so scared to read the email. What I did instead is looking for any 'happy words' such as 'delightful', 'congratulations', 'pleased', etc. But you know what? I didn't find any of these words but I found the word 'regret'! Oh Shit!!! as soon as I saw this I said to myself this is another unlucky paper. Oh!!! I was really disappointed and I can feel my blood was boiling that time. I can feel my vein kicking and pumping off my red blood to my brain which caused millions of questions. For the first few minutes, it was really really hard to accept the fact that my paper has been rejected (after my ICPR paper has been rejected recently). I was so fucking upset but after half an hour my blood pressure went down. I think if I had measured my blood pressure in the first a few minutes it went up to 180/90 which is very very dangerous (I could be dead after reading the email LOL!).

I read the feedback from the reviewers. The same to ICPR, one reviewer accepted it and the other two rejected it. Next I read every single line of their comments to make sure I am wrong in some parts of my paper. I agreed with some of their comments (maybe about 20%) but the other 80% I don't think their comments are helpful. In fact, some of the comments show how naive he/she is in image processing (am I being over the top now? probably...but until you've seen this one particular comment then you understand me LOL!). Anyway, on the good side of his/her comments, he/she suggested me to do one thing which is to normalise my histograms. Initially I disagreed because histogram normalisation is implemented when we want to measure similarity between two histograms. But I persuaded my 'black heart' to just bloody do it and see how's the results. I took about 10 or 15 minutes to modify my algorithm and ran the program. After about 3 hours I got the results...ta..da...............surprisingly accuracy, sensitivity and specificity have increased 4% which makes my method produced almost 90% accuracy. But then it was only 4% and I'm still not sure how significant the increments are. But the main thing is..it is visually better to present 89% sensitivity instead of 85%.

After a long discussion with my supervisor we agreed that the reviewers did not read thoroughly my paper. They probably jumped from the first line to the third line and to the seventh line. If this is the way they did it then I would say I hope their papers will be rejected as well in other journals or conferences (wow...is this me? swearing LOL!). If my ICPR rejected paper is submitted to Portugal, my MIUA rejected paper will be submitted to Ireland. Well..well...well....what do you know? there might be reasons behind of double rejection? God knows! In a couple of weeks time I will be going to London. I hope that I can present my ICCCV paper fluently and I hope London brings some luck to me :-)......Summer is around the corner and I can wait to wear my shorts and t-shirts again. Phewwww.....I am tired I need a break either in June or July.

No comments:

Other stories

Related Posts with Thumbnails