Navdeep Kanwal

Punjabi University, Patiala · Department of Computer Engineering

What are the journal available in computer science field to publish excessive long survey papers with approximate page length of 60 to 70 pages in single column and appx 40 pages in double column?

Any suggestions for name of journals which are publishing research related to solutions or surveys for future smart cities or smart city models.

Digital Image processing has been a field of interest for the the computer science researchers. What can be the different areas of digital image processing, in which a researcher should start its research ?

3D Imaging?

Video surveillance?

Digital Image Forensics?

or some other field in which we can deliver something promising?

I am a newcomer to Machine learning and classification. I am working for classification of two different type of image classes. I have calculated block-wise(overlapping,sliding(1px), size:3x3) features of every image in the dataset and stored them row wise. There are 800 images in one type of class (A) of dataset and 450 images in other type of class (B) in the dataset. so, basically i have an array of 800 rows and 90000(appx) column in one class and 450 rows and 90000(appx) column in other class.

now i want to train and test SVM classifier on these two classes. I have tried to perform 10 fold classification using some of following methods:

1. every alternative image from class A and first 400 from class B accuracy was around 70% (low TP, positive i am considering for minor class B)

2. all images from class A and all from class B, accuracy was around 80% average

3. in third case i have upsampled class B with basic repetition measure i.e all images from class A, all images from B + first 350 images again from class B which boosted accuracy to above 90% with high TP and high TN

i am using SVM as following:

SVMModel = fitcsvm(trained_data,gg,'Standardize',true,'KernelFunction','RBF','BoxConstraint', 32, 'KernelScale', 0.2008);

although i am getting nice accuracy but i am not sure about acceptability of my upsampling methodology.

question 1: am i using right method to upsample the feature set? if not what other suitable method i can use? any suggestions. (I have already tried SMOTE but it doesnt work because my data has very small values standard deviation and SMOTE perform addition/multiplication operation which causes major changes in the data, shifting to some other type class)

question 2: can i use SVM without manual upsampling in this case? If yes then what should i write in place of following to tune SVM?

SVMModel = fitcsvm(trained_data,gg,'Standardize',true,'KernelFunction','RBF','BoxConstraint', 32, 'KernelScale', 0.2008);

well, kernal scale i have decided according to misclassfication rate on crossvalidation (third case).

please guide me.....thanks in advance.

I belong to Computer Science and Engineering field. I have got a journal SOIC listed in scopus as well as indexed by Mathematical Reviews/MathScinet.

As SCI/SCIE is known as good indexing agency in engineering/ sciences streams, I want to know about the value and reputation of Mathematical Reviews/Mathscinet indexing. The jounal is highlighting that over scopus. May be somebody related to mathematics field may answer...Should I consider Mathematical Reviews/Mathscinet (American Mathematical Society) as a good indexing agency for the journal as of SCI/SCIE??

regards

I am trying to understand and then implement zernike moments for overlapping blocks in an image. I have read that, to compute the Zernike moments of a given block, the center of the block is taken as the origin and pixel coordinates needs to be mapped to the range of the unit circle.

How will we map the pixel coordinates of a centre pixel to the unit circle range. can anybody please elaborate by some example or some link for the same.

using following standard equations of zernike for the computation:

I want to perform a block by block comparison by using overlapping blocks and their respective image moments.

Can anybody suggest for type of image moment suited for blocks of DCT/DWT of that image. Hu moments are working fine for uint8 image but do not perform for DCT/DWT of same image.in DWT i wish to work on LL band.

need to submit a Survey paper with theoretical comparisons with concluding summary, in the field of Image Processing/Image Forensics

I have an array containing thousands (may be lakhs) of binary values. Want to apply bit shifting operations on that. bitshift,bitsra etc. works on integer values in matlab and it is not possible to convert a binary number of 10-12 thousand bits to decimal, as it will be infinity. Any suitable solution???

Does MATLAB support zero based indexing

some program needs the logic to be driven through that