A Novel Biometric Image Enhancement Approach With the Hybridization of Undecimated Wavelet Transform and Deep Autoencoder

For a long time, image enhancement techniques have been widely used to improve the image quality in many image processing applications. Recently, deep learning models have been applied to image enhancement problems with great success. In the domain of biometric, fingerprint and face play a vital role to authenticate a person in the right way. Hence, the enhancement of these images significantly improves the recognition rate. In this chapter, undecimated wavelet transform (UDWT) and deep autoencoder are hydridized to enhance the quality of images. Initially, the images are decomposed with Daubechies wavelet filter. Then, deep autoencoder is trained to minimize the error between reconstructed and actual input. The experiments have been conducted on real-time fingerprint and face images collected from 150 subjects, each with 10 orientations. The signal to noise ratio (SNR), peak signal to noise ratio (PSNR), mean square error (MSE), and root mean square error (RMSE) have been computed and compared. It was observed that the proposed model produced a biometric image with high quality.

Reduced Cycle Spinning Method for the Undecimated Wavelet Transform

The Undecimated Wavelet Transform is commonly used for signal processing due to its advantages over other wavelet techniques, but it is limited for some applications because of its computational cost. One of the methods utilized for the implementation of the Undecimated Wavelet Transform is the one known as Cycle Spinning. This paper introduces an alternative Cycle Spinning implementation method that divides the computational cost by a factor close to 2. This work develops the mathematical background of the proposed method, shows the block diagrams for its implementation and validates the method by applying it to the denoising of ultrasonic signals. The evaluation of the denoising results shows that the new method produces similar denoising qualities than other Cycle Spinning implementations, with a reduced computational cost.

Morphological Component Analysis for Biological Signals

This chapter explains the removal of artifacts from the multi-resource biological signals. Morphological components can be used to distinguish between the brain activities and artifacts that are contaminated with each other in many physical situations. In this chapter, a two-stage wavelet shrinkage and morphological component analysis (MCA) for biological signals is a sophisticated way to analyze the brain activities and validate the effectiveness of artifacts removal. The source components in the biological signals can be characterized by specific morphology and measures the independence and uniqueness of the source components. Undecimated wavelet transform (UDWT), discrete cosine transform (DCT), local discrete cosine transform (LDCT), discrete sine transform (DST), and DIRAC are the orthonormal bases function used to build the explicit dictionary for the decomposition of source component of the biological signal in the morphological component analysis. The chapter discusses the implementation and optimization algorithm of the morphological component analysis.