Multifocus Color Image Fusion Based on Walsh Hadamard Transform and SumModified-Laplacian Focus Measure

Abstract

Multi-focus image fusion objective is to add relevant information from multiple images of the same scene but with different focuses into a sharper image that is more suitable for visual sensor networks. Natural and artificially obtained multifocus color images are considered for fusion. The existing fusion methods like multi scale and multi-resolution transforms are proved to be good in Multi-focus Image Fusion. However, they suffer from computational complexity in kernels calculation. In this paper, Multi-focus color Image Fusion based on Walsh-Hadamard Transform and sum-modified-Laplacian focus measure is proposed. Walsh-Hadamard Transform is a non-sinusoidal, orthogonal transform with symmetry, seperability and orthogonal properties. These properties make it more apt for image fusion than other transforms. And the sum-modified-Laplacian focus measure helps to get sharper image. Proposed method performance is evaluated in terms of reference and non-reference measures. The experimental results indicate that proposed method not only produces sharp details in fused image but also reduces the computational complexity.

Hybrid Multi-Focus Image Fusion using Stationary Wavelet Transform and Focus Measures For Visual Sensor Networks

“…Natural and misregistered image fusion performance is evaluated in terms of non-reference measures Q BC [18] and Q Y [19]. Table I and Table II gives the comparison of proposed fusion method performance with DCHWT [3], MSVD [6], SWT [8], DDDWT [9], DTCWT [10] and WHT [5] methods in terms of aforesaid performance measures. [6] 27.7903 0.9335 SWT [8] 28.2727 0.9374 DDDWT [9] 28.2718 0.9372 DTCWT [10] 32.9903 0.9880 WHT [5] 33.5581 0.9962 Proposed method 33.8149 0.9974 Airplane DCHWT [3] 33.0324 0.9843 MSVD [6] 26.3356 0.9173 SWT [8] 27.2888 0.9253 DDDWT [9] 27.3447 0.9238 DTCWT [10] 32.5604 0.9875 WHT [5] 33 [6] 0.7711 0.9203 SWT [8] 0.7733 0.9344 DDDWT [9] 0.7716 0.9391 DTCWT [10] 0.7778 0.9270 WHT [5] 0.8102 0.9565 Proposed method 0.8121 0.9566 Children DCHWT [3] 0.7246 0.9376 MSVD [6] 0.7056 0.8304 SWT [8] 0.7137 0.9230 DDDWT [9] 0.7113 0.9286 DTCWT [10] 0.7248 0.9345 WHT [5] 0 Fig.…”

“…To estimate the performance and robustness of our method, we also compared the fusion results of misregistered images in Table III. [6] 0.5618 0.8439 SWT [8] 0.5634 0.8701 DDDWT [9] 0.5596 0.8804 DTCWT [10] 0.6298 0.9102 WHT [5] 0…”

“…To address these issues, fusion techniques are desirable for fusing two or more images with divergent focus levels into a focused fused image [1]. In literature, various image fusion techniques were developed like Gradient pyramids [2], Discrete Cosine Harmonic Wavelet Transform (DCHWT) [3], discrete cosine transform [4], Walsh Hadamard Transform [5], multiresolution singular value decomposition (MSVD) [6], Wavelet based methods [7][8][9][10]. The discrete wavelet transform (DWT) based method had been verified to be an effective image fusion technique.…”

“…Focus measures used in this proposed algorithm are given below. Sum-modified-Laplacian (SML): It is a focus measure used to select focused areas from source images to build the fused image with each object in focus. SML focus measure based on Laplacian of the image and its mathematical background is available in[5]. Compute the SML of Each block of high-frequency sub-bands and the sub-bands with high sum modified Laplacian measure are used to obtain the high frequency part of fused image.…”

Stationary Wavelet Transform based Image Fusion using fusion rules

“…The fusion techniques using Laplacian pyramids [2], Discrete Wavelet Transform (DWT) [3], discrete cosine transform [4], Walsh Hadamard Transform [5], multiresolution singular value decomposition (MSVD) [6], Wavelet based methods [7][8][9][10][11][12] are existing in the literature. The discrete wavelet transform (DWT) based method had been verified to be an effective image fusion technique.…”