A multi-focus image fusion method via region mosaicking on Laplacian pyramids

Liang, Kun; Zhang, Liguo; Zhang, Kejia; Sun, Jun; Han, Qilong; Jin, Zilong

doi:10.1371/journal.pone.0191085

Cited by 35 publications

(19 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Different transform domain methods are used for image resolution enhancement through the transformation of the source image into different scales, which then are composed into one fused image [ 12 ]. Commonly used techniques are wavelet, curvelet and contourlet transforms, neighbour distance, Laplacian pyramid or gradient pyramid [ 13 , 14 ]. Deep learning methods (specifically CNN based approaches) are often incorporated to solve blurring-effect problems through the ability to learn the focus measure to recognize the focused and defocused pixels or regions in source images [ 15 , 16 ]; to learn the fusion operation to fuse a pair without the need for ground truth fused images [ 17 , 18 ]; to learn the direct mapping between the high-frequency and low-frequency images of the source and fusion images [ 19 ], and so forth.…”

Section: Related Workmentioning

confidence: 99%

“…The performance time estimated for 10 images of 320 × 240 resolution is 4.6 s using a robust sparse representation model with a Laplacian regularization, but the time jumps up to 31.6 s if the stock is increased to 20 images [ 23 ]. For microscopic images, some techniques allow reducing the processing time to 1.35 s (using seven layers Laplacian pyramid for 720 × 480 images) [ 13 ].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Fast Multi-Focus Fusion Based on Deep Learning for Early-Stage Embryo Image Enhancement

Raudonis

Paulauskaitė-Tarasevičienė

Šutienė

2021

Sensors

View full text Add to dashboard Cite

Background: Cell detection and counting is of essential importance in evaluating the quality of early-stage embryo. Full automation of this process remains a challenging task due to different cell size, shape, the presence of incomplete cell boundaries, partially or fully overlapping cells. Moreover, the algorithm to be developed should process a large number of image data of different quality in a reasonable amount of time. Methods: Multi-focus image fusion approach based on deep learning U-Net architecture is proposed in the paper, which allows reducing the amount of data up to 7 times without losing spectral information required for embryo enhancement in the microscopic image. Results: The experiment includes the visual and quantitative analysis by estimating the image similarity metrics and processing times, which is compared to the results achieved by two wellknown techniques—Inverse Laplacian Pyramid Transform and Enhanced Correlation Coefficient Maximization. Conclusion: Comparatively, the image fusion time is substantially improved for different image resolutions, whilst ensuring the high quality of the fused image.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Fast Multi-Focus Fusion Based on Deep Learning for Early-Stage Embryo Image Enhancement

Raudonis

Paulauskaitė-Tarasevičienė

Šutienė

2021

Sensors

View full text Add to dashboard Cite

show abstract

“…First, we take the input image for downsampling and Gaussian blur at different scales and establish multiple sets of multi-scale space sequences to form the image Gaussian pyramid. Then we subtract adjacent images in each set of scale-space sequences, forming a difference of Gaussian pyramid(DOG) [44].…”

Section: Image Registrationmentioning

confidence: 99%

Dual-Enhanced Registration for Field of View Ultrasound Sonography

et al. 2020

View full text Add to dashboard Cite

Extended Field of View Ultrasound Sonography (EFOV-US) uses the existing ultrasound images for image stitching, so as to display the shape and scope of organ occupation and the relationship with surrounding tissues comprehensively. However, there are still some problems in Extended Field of View Ultrasound Sonography, such as matching error and unstable quality of image stitching. In view of these problems, we propose Dual-enhanced EFOV-US method that overcomes the limitation and produces higher quality results. Firstly, the gray enhancement method is used to improve the image contrast and reduce the noise interference. Then the super-resolution method based on the generative adversarial network is used to improve the resolution of the ultrasonic image further and increase the number of feature point matching between stitching images. The high quality ultrasound wide-range image is gotten by stitching and fusing the double enhanced image. The experimental results show that the proposed method is effective and practical.

show abstract

“…Generally, the decomposition procedure is completed with a Multi-Scale Transform (MST) [9]. Typical transform algorithms such as Wavelet Transform (DWT) [10], Laplacian Pyramid (LP) [11], Curvelet Transform (CvT) [12], Contourlet Transform (CT) [13], Stationary Wavelet Transform [14], Nonsubsampled Contourlet Transform [15] and Nonsubsampled Shearlet Transform [16] are MST image transformations that decompose the source images into a set of sub-images to extract the detailed information. After the decomposition, the fused sub-images (coefficients) are obtained by specific fusion rules as Choosing Max (CM) and Weighted Average (WA) [17][18][19].…”

Section: Introductionmentioning

confidence: 99%

Multi-Focus Image Fusion with Multi-Scale Transform Optimized by Metaheuristic Algorithms

Abas¹,

Baykan²

2021

View full text Add to dashboard Cite

Focus is limited and singular in many image capture devices. Therefore, different focused objects at different distances are obtained in a single image taken. Image fusion can be defined as the acquisition of multiple focused objects in a single image by combining important information from two or more images into a single image. In this paper, a new multi-focus image fusion method based on Bat Algorithm (BA) is presented in a Multi-Scale Transform (MST) to overcome limitations of standard MST Transform. Firstly, a specific MST (Laplacian Pyramid or Curvelet Transform) is performed on the two source images to obtain their low-pass and high-pass bands. Secondly, optimization algorithms were used to find out optimal weights for coefficients in low-pass bands to improve the accuracy of the fusion image and finally the fused multi-focus image is reconstructed by the inverse MST. The experimental results are compared with different methods using reference and non-reference evaluation metrics to evaluate the performance of image fusion methods.

show abstract

A multi-focus image fusion method via region mosaicking on Laplacian pyramids

Cited by 35 publications

References 45 publications

Fast Multi-Focus Fusion Based on Deep Learning for Early-Stage Embryo Image Enhancement

Fast Multi-Focus Fusion Based on Deep Learning for Early-Stage Embryo Image Enhancement

Dual-Enhanced Registration for Field of View Ultrasound Sonography

Multi-Focus Image Fusion with Multi-Scale Transform Optimized by Metaheuristic Algorithms

Contact Info

Product

Resources

About