Comparative Study of Vector Quantization in Image Compression
Abstract
Image Compression holds a technique of compressing the data which encodes the standard image with fewer amount of bits, which results from particular sampling values using quantization methods. Quantization is a lossy compression technique, which is done by minimizing a set of bits into single quantum bit. In Vector Quantization, codebook is the significant part. The various techniques for quantization are explained in this paper. Some popular quantization techniques consume less memory and computation time such as LBG80, FBP, BPNN, ENN and Simple nearest neighbor algorithm. The resultant values of different quantization are explained in an experimental part, with the use of standard performance measure of MSE and PSNR. This approach gives an effective Codebook with minimum computational time and includes excellent Peak Signal to Noise Ratio values (PSNR).
About this article
Keywords
- Compression
- Transformation - Discrete Cosine Transformation (DCT)
- Discrete Wavelet Transformation (DWT)
- Quantization-LBG algorithm.
References
[1] Karthikeyan N, “A new KK coder with Low Computation and Low Memory for Image Compression” , International Journal, Vol-9 No.27, May 20,2016
[2] Karthikeyan N and Sivakumar N, “Comparative Study of Different Encoder Schemes of DCT for Image Compression”, International Journal of Applied Engineering and Research, Volume 10, Number 49, PP 135- 139, 2015.
[3] Mukesh Mittal, RuchikaLamba” Image Compression Using Vector Quantization Algorithms”, International Journal, Volume-3, Issue 6, June 2013
[4] Wei-Yi Wei Graduate Institute of Communication Engineering National Taiwan University” An Introduction to Image Compression
[5] Sayan Nag, “Vector Quantization using the Improved Differential Evolution Algorithm for Image Compression”, on 15 October 2017.
[6] K.Somasundaram, and S.Domnic, “Modified Vector Quantization Method for Image Compression”, vol-13, May 2006.
[7] Jau-JiShen,” A New Approach of Image Compression Based on Difference Vector Quantization”,2011
[8] Ya-Hsin Lo, “A New Approach of Image Compression Based on Difference Vector Quantization”
[9] YosephLinde, AndresBuzo, RobertM.Gray,”An Algorithm for Vector QuantizerDesig” ,IEEE Transaction, vol. com-28, no. 1, january 1980
[10] LecturerCornelBalint, ”An improved LBG algorithm for vector quantization”,Anale.SerialInformatica.vol.II,face.I
[11] Ms.AsmitaA.Bardekar, Mr.P.A.Tijare,” A Review on LBG Algorithm for Image Compression”, IJCSIT Vol. 2 (6) , 2011
[12] ]Arup Kumar Pal and AnupSar,”An Efficient Codebook Initialization Approach For Lbg Algorithm “,IJCSEAVol.1, No.4, August 2011
[13] AsifUllah Khan, “Genetic Algorithm Based Backpropagation Neural Network Performs better than Backpropagation Neural Network in Stock Rates Prediction”, IJCSNS ,VOL.8 No.7, July 2008
[14] T. K. Bandopadhyaya, ”Genetic Algorithm Based Backpropagation Neural Network Performs better than Backpropagation Neural Network in Stock Rates Prediction”IJCSNS , VOL.8 No.7, July 2008
[15] Sudhir Sharma, ”Genetic Algorithm Based Backpropagation Neural Network Performs better than Backpropagation Neural Network in Stock Rates Prediction”IJCSNS , VOL.8 No.7, July 2008
[16] NazriMohd.Nawi, Abdullah Khan, M. Z. Rahman” A New Back-Propagation Neural Network Optimized with Cuckoo Search Algorithm”
[17] Karthikeyan N, Sivakumar N and Venkatesh V “A Fast and Efficient Lifting Based DCT Image Compression”, Pakistan journal of Biotechnology, Vol.13, Pp. 544- 547, 2016.
[18] K. Vijayakumar C. Arun, Continuous security assessment of cloud based applications using distributed hashing algorithm in SDLC, Cluster Computing DOI 10.1007/s10586-017-1176-x,Sept 2017
[19] R.Joseph Manoj, M.D.Anto Praveena, K.Vijayakumar, “An ACO–ANN based feature selection algorithm for big data”, Cluster Computing The Journal of Networks, Software Tools and Applications, ISSN: 1386-7857 (Print), 1573-7543 (Online) DOI: 10.1007/s10586-018-2550-z, 2018.