Baisa L. Gunjal

Work place: Amrutvahini College of Engineering, Sangamner, Ahmednagar, MS, India



Research Interests: Image Compression, Computer Networks, Network Architecture, Image Processing


Dr. Baisa L. Gunjal has completed PhD from Savitribai Phule Pune University and working as associate professor in Amrutvahini College of Engineering, Sangamner, Ahmednagar, Maharashtra, India. She is recipient of six awards during 2012 to 2015. She is recipient of ‘Maximum Publications award-2013’ from Computer Society of India, ‘Best Teacher Award-2013’ from Savitribai Phule Pune University(SPPU), MS, India, ‘Lady Engineer Award-2012’ from Institution of Engineers, ‘Active faculty Award-2012’ from Computer Society of India, ‘Best Research Paper Award’ in 11th IEEE Indian International Conference (INDICON) in 2014 and ‘Best Research Paper Award in international conference. Technovision-2014, MS India. She has more than 25 international journals and conference publications including SpringerPlus, IEEE Computer society, ACM digital library, IET digital library, Computer Society of India Communications, World Academy of Science, Engineering and Technology (WASET). She is author of book “Software Engineering” written for undergraduate engineering students of SPPU and completed watermarking based research project funded by board of colleges and university development (BCUD) of SPPU in 2013. She has got copyright from government of India for watermarking based research work in 2014. Her areas of interest include image watermarking, network security, data structures and advanced databases.

Author Articles
Robust, Secure and High Capacity Watermarking Technique based on Image Partitioning-Merging Scheme

By Baisa L. Gunjal

DOI:, Pub. Date: 8 Apr. 2016

This paper presents secure and high capacity watermarking technique using novel approach of Image Partitioning-Merging Scheme (IPMS). The IPMS is used as reduction method to reduce the size of watermark logically and increases security levels of proposed watermarking technique. The technique effectively uses special properties of Discrete Wavelet Transform (DWT), Fast Walse Hadamrd Transform (FWHT), Singular Value Decomposition (SVD) and proved strongly robust to 14 noise addition and filtering attacks. Fibonacci Lucas Transform (FLT) is used as effective scrambling technique to scramble the watermark to provide additional security in embedding process. Many researchers failed to achieve imperceptibility and robustness under high watermark embedding scenario with strong security provision as these quality parameters conflict each other. The novel technique presented here archives imperceptibility, high capacity watermark embedding, security and robustness against 14 noise addition and filtering attacks. The technique is non-blind and tested with grey scale cover images of size 512x512 and watermark images of size 512x512. The experimental results demonstrate that Lena image gives 75.8446dBs imperceptibility which is measured in terms of Peak signal to noise ratio. The robustness is measured in terms of normalized correlation (NC) equals to 1 showing exact recovery of watermark. The method is found strongly robust against noise addition and filtering attacks with compared to existing watermarking methods under consideration.

[...] Read more.
Strongly Robust and Highly Secured DWT-SVD Based Color Image Watermarking: Embedding Data in All Y, U, V Color Spaces

By Baisa L. Gunjal Suresh N. Mali

DOI:, Pub. Date: 8 Apr. 2012

In this paper ‘DWT-SVD’ based Color Image Watermarking technique in YUV color space using Arnold Transform is proposed. The RGB color image is converted into YUV color space. Image is decomposed by 3 level DWT and then SVD is applied. The security is increased with watermark scrambling using Arnold Transform. The watermark is embedded in all Y,U and V color spaces in HL3 region. The decomposition is done with ‘Haar’ which is simple, symmetric and orthogonal wavelet and the direct weighting factor is used in watermark embedding and extraction process is used. PSNR and Normalized Correlations (NC) values are tested for 10 different values of flexing factor. We got maximum PSNR up to 52.3337 for Y channel and average value of NC equal to 0.99 indicating best recovery of watermark. The proposed scheme is non blind and strongly robust to different attacks like compression, scaling, rotation, cropping and Noise addition which is tested with standard database image of size 512x512 and watermark of size 64X64.

[...] Read more.
Other Articles