Browsing by Author "Brar, Sukhreet Singh"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    DWT-SVD Based robust digital image watermarking using adaptive median filter
    (Central University of Punjab, 2013) Kumar, Ashok; Brar, Sukhreet Singh
    Feature Matching is one of the central issues of model-based recognition and an important component of most object recognition systems. The feature matching technique can be applied in image alignment, 3D secure reconstruction, motion tracking, object recognition and in pharmacophore based molecular alignment. Even though numerous algorithms exist for feature recognition and search yet the efficiency of these have considerable room for improvement. This has necessitated a continuous search for better performing algorithms. The combination of geometric hashing technique and tolerance limits for a given set of known interactions can be used for the pharmacophore modelling based search. The algorithm proposed is precise as compared with the previous existing algorithms. This algorithm is better in its runtime complexity.
  • Thumbnail Image
    Item
    Video quality evaluation using DWT-SPIHT based watermarking technique
    (Institute of Electrical and Electronics Engineers Inc., 2018) Kumar, Rajeev; Kumar, Sanjeet; Brar, Sukhreet Singh
    The video quality analysis is very important in today's video broadcasting and transmission quality control because the quality is a key factor in customer satisfaction. This paper presents a full reference based video quality evaluation technique, using the application of watermarking. Frames with low complexity value are selected for the insertion of the watermark, on the selected video frames; DWT is applied up to third level to get the DWT coefficients. Then SPIHT encoding/decoding is applied which helps in finding the significant coefficients for the insertion of watermark. The significant coefficients are converted into bit-planes and 3rd bit plane (from LSB side) of each coefficient is selected for watermark. The SPIHT algorithm ensures that the watermark bits are inserted uniformly throughout the frames. After embedding the watermark, the IDWT is applied and we get the watermarked frames which are then incorporated in the video file. The simulation results show no visible difference between watermarked frame and original frame. The video quality is estimated in terms of peak signal to noise ratio (PSNR), structure similarity (SSIM) and true detection rates (TDR) against various attacks like Salt pepper noise, rotation Gaussian noise and compression attacks. The results indicate that the proposed technique for quality evaluation of the video, accurately calculates the quality of a video and also it is robust against various attacks. ? 2016 IEEE.