Focus Image Fusion Research Articles

A Detailed Survey of Multi Focus Image Fusion

A Detailed Survey of Multi Focus Image Fusion

An autoencoder deep residual network model for multi focus image fusion

An autoencoder deep residual network model for multi focus image fusion

Continuous optical zoom microscope with extended depth of field and 3D reconstruction

Microscope such as fluorescence microscope, confocal microscope and two-photon microscope plays an important role in life science, laser processing and other fields. However, most microscopes only have discrete zoom rates. In this paper, a continuous optical zoom microscope with extended depth of field and 3D reconstruction is demonstrated for the first time. It consists of a zoom objective lens, a microscope holder, an adjustable three-dimensional object stage, an Abbe condenser and an LED light source. The zoom objective lens is composed of several liquid lenses and solid lenses. By adjusting the applied voltage to the liquid lens, the proposed microscope can achieve a large continuous magnification from 10? to 60?. Moreover, an improved shape from focus (SFF) algorithm and image fusion algorithm are designed for 3D reproduction. Based on the liquid lenses, the axial focusing position can be adjusted to obtain images with different depths, and then the extended depth of field and 3D reconstruction can be realized. Our experimental results demonstrate the feasibility of the proposed microscope. The proposed microscope is expected to be applied in the fields of pathological diagnosis, biological detection, etc.

A New Robust Multi focus image fusion Method

In today's digital era, multi focus picture fusion is a critical problem in the field of computational image processing. In the field of fusion information, multi-focus picture fusion has emerged as a significant research subject. The primary objective of multi focus image fusion is to merge graphical information from several images with various focus points into a single image with no information loss. We provide a robust image fusion method that can combine two or more degraded input photos into a single clear resulting output image with additional detailed information about the fused input images. The targeted item from each of the input photographs is combined to create a secondary image output. The action level quantities and the fusion rule are two key components of picture fusion, as is widely acknowledged. The activity level values are essentially implemented in either the "spatial domain" or the "transform domain" in most common fusion methods, such as wavelet. The brightness information computed from various source photos is compared to the laws developed to produce brightness / focus maps by using local filters to extract high-frequency characteristics. As a result, the focus map provides integrated clarity information, which is useful for a variety of Multi focus picture fusion problems. Image fusion with several modalities, for example. Completing these two jobs, on the other hand. As a consequence, we offer a strategy for achieving good fusion performance in this study paper. A Convolutional Neural Network (CNN) was trained on both high-quality and blurred picture patches to represent the mapping. The main advantage of this idea is that it can create a CNN model that can provide both the Activity level Measurement" and the Fusion rule, overcoming the limitations of previous fusion procedures. Multi focus image fusion is demonstrated using microscopic images, medical imaging, computer visualization, and Image information improvement is also a benefit of multi-focus image fusion. Greater precision is necessary in terms of target detection and identification. Face recognition" and a more compact work load, as well as enhanced system consistency, are among the new features.

DMDN: Degradation model-based deep network for multi-focus image fusion

Multi-focus image fusion (MFIF) is an efficient technique that merges differently focused images into one all-in-focus image, breaking the focus limitation of camera imaging. Most of the existing deep learning-based MFIF methods follow the fusion after focus measure strategy that divides the MFIF problem into focus map generation and image fusion. However, the generated focus map is usually coarse and the quality of the fused image greatly depends on post-processing. Different from the fusion after focus measure strategy, we propose a mathematical degradation model, where we assume the defocused images are degraded from a latent all-in-focus image, and the approximated all-in-focus image can be directly estimated from the degraded images. Thus we construct an end-to-end deep neural network to learn the mapping between the defocused images and the all-in-focus images. During training, the network inputs a pair of clear image and global blurred image in a random concatenate order, while the output fused image is supervised by the clear image with a Mix-loss comprising the pixel-wise loss and the perceptual loss. Our method is compared with the state-of-the-art algorithms on two MFIF datasets and our newly published dataset. Experimental results demonstrate the effectiveness of our method subjectively and objectively.

Image fusion algorithm based on unsupervised deep learning-optimized sparse representation

The image fusion method based on deep learning has problems such as the supervised learning of the model, the edge and noise of the fused image, and the setting of the image fusion weight map. To solve these problems, this paper proposes an end-to-end unsupervised deep learning model that performs one-to-many focus image fusion. It solves the training problem encountered by supervised deep learning models while avoiding the unreasonable image fusion weight maps. In addition, this paper proposes an optimized sparse representation method that divides an image into a target area and a background area. Then, it uses super complete dictionary learning to obtain a sparse representation of the image background area. This approach makes the proposed unsupervised deep learning image fusion method robust to noise. Finally, using this method to carry out image fusion experiments, the results show that the quality evaluation indicators of the fused image obtained by this method substantially outperform those of both mainstream machine learning and deep learning image fusion methods.

Multi Focus Image Fusion with Region-Center Based Kernel

The usage of the camera is ubiquitous nowadays. It provides highly accurate information. Then, the camera helps humans to carry out their specific tasks correctly. Furthermore, the camera becomes an important tool to achieve accurate computation in some fields, such as in medical diagnostic, robotics, remote sensing, and others. On the other side, the camera also has a weakness to capture detailed information of the scene in one image. Many images are needed to obtain the focus information of all the scenes since the lens's limitation depth of field produces out of focus regions beyond the focused object. To make a settlement of that case, the researchers have found a multi-focus image fusion process. This process selects all detailed information from a sequence of images and fuses them into one focused image. Through this fused image, the user such as human and machine can read the focus information easier. Later, the researchers developed multi-focus image fusion methods which various advanced procedures and algorithms. Furthermore, the implementation of multi-focus image fusion in new fields multiplied in the last two decades ago. It was able to create an accurate and efficient method to build a fused image. The proposed method is a kind of a new method in multi-focus image fusion. It works according to the region-center based kernel. The kernel processes input image to predict the detailed information of the scene. This method is robust to prevent the unexpected effects and sensitivities of noise. The proposed method generates a fused image with high accuracy. Finally, the assessment is done based on mutual information and structure similarity.

Renyi entropy and atom search sine cosine algorithm for multi focus image fusion

This paper proposes a novel strategy, atom search sine cosine algorithm (ASSCA) for multi-focus image fusion. Here, the discrete wavelet transform (DWT) is adapted for transforming images into sub-bands. The fusion is carried out using a fusion rule based on weighting criteria that uses two attributes, Renyi entropy and the proposed ASSCA. Entropy discovers the entropy fusion factor considering the assessed entropy from the source image. Further, an optimization strategy, ASSCA is developed by integrating atom search optimization and sine cosine algorithm for precise selection of fusion factor. The output obtained from the fusion undergoes inverse discrete wavelet transform to obtain the resultant fused image. The proposed DWT + ASSCA + Renyi entropy outperformed other methods with maximal mutual information of 1.492, maximal peak signal-to-noise ratio of 40.625 dB, minimal root mean-squared error of 7.651, maximum correlation coefficient of 0.988, and minimum deviation index of 1.146.

Multifocus image fusion using te-cnn technique

A combination of two or more images is called an image fusion which produce an image that teaches more about computer perspective or human perception. The multi-focus image fusion technique in this paper uses CNN (Convolutional Neural Networks) to create a focus pixel-level maps on dynamic neural networks from multicolored images. This proposed CNN technique is entirely new approach when compared to the other integration methods which are utilized the same CNN method. In this proposed approach, the binary division is completely accessed to generate the focus maps. The result of this implementation prepares high-resolution pure images, view maps using bright field data. Also, produces exact focus maps next to a clear visual image. These mixing methods produce a proportional measurement. The deep learning method is used to perform the pixel separation for multi focus image fusion technique. The deep learning algorithm creates the datasets which are necessary to perform the image fusion in multi focus images by using focus maps and focus images.

An Improved Multi Focus Image Fusion Method Using Non Sub-sampled Contourlet Transform and Guided Filter

An Improved Multi Focus Image Fusion Method Using Non Sub-sampled Contourlet Transform and Guided Filter

A novel multi-focus image fusion paradigm: A hybrid approach

This article discusses a hybrid wavelet-based multi- focus image fusion technique using Principal Component Analysis (PCA). Previously many authors have used many wavelet transforms in a different way for effective image fusion purpose but in this article, the authors have used Discrete Wavelet Transform (DWT) and Stationary Wavelet Transform (SWT) to process the input images in parallel. Later their results are further merged/ fused using standard PCA method. The results of the proposed method are compared with some of the conventional and non-conventional works related to SWT, DWT, and PCA. The result shows a good result than the compared method’s result. The results are analyzed and assessed using visual appearance of the final fused image and also using some of the parameters like Standard Deviation (SD), Petrovics metric (PM), Average Pixel Intensity (API) and Average Gradient (AG).

Multi Focus Image Fusion Based on Non Subsampled Contour & Focused Area Detection

Multi Focus Image Fusion Based on Non Subsampled Contour & Focused Area Detection

Image Fusion Based Multimodal Biometric Recognition

Biometric Authentication is a security process that replays on the unique biological characteristics of an individual. Biometric Authentication system compare a biometric data capture to stored, confirmed authentic data in a database. It is simply the process of verifying the identity using the measurements or other unique characteristics of the body, then logging us in a service, device and so on. It is an effective way to prove identity because it can’t be replicated. Multi focus Image fusion is a process of fusing two or more images to obtain a new one. Used to reduce the problems like blocking, ringing artifacts occurs because of DCT. The low frequency sub-band coefficients are fused by selecting coefficient having maximum spatial frequency. The goal is classifying the images to classes of authorized and unauthorized using multi class SVM. The fingerprint image and iris image are fused together using SWT, the features are extracted from the fused image and labelled using GLCM algorithm. The testing image is then compared with trained samples and classified as authorized or unauthorized by using FFNN.

Multi-Focus Image Region Fusion and Registration Algorithm with Multi-Scale Wavelet

Aiming at the problems of poor brightness control effect and low registration accuracy in traditional multi focus image registration, a wavelet multi-scale multi focus image region fusion registration method is proposed. The multi-scale Retinex algorithm is used to enhance the image, the wavelet decomposition similarity analysis is used for image interpolation, and the EMD method is used to decompose the multi focus image. Finally, the image reconstruction is completed and the multi focus image registration is realized. In order to verify the multi focus image fusion registration effect of different methods, a comparative experiment was designed. Experimental results show that the proposed method can control the image brightness in a reasonable range, the root mean square error of image region fusion registration algorithm is less than 5, and the image registration accuracy is high. This method can achieve multi focus image region fusion registration.

Multi -Focus Image Fusion Based on Stationary Wavelet Transform and PCA on YCBCR Color Space

The multi-focus image fusion method can fuse more than one focused image to generate a single image with more accurate description. The purpose of image fusion is to generate one image by combining information from many source images of the same scene. In this paper, a multi-focus image fusion method is proposed with a hybrid pixel level obtained in the spatial and transform domains. The proposed method is implemented on multi-focus source images in YCbCr color space. As the first step two-level stationary wavelet transform was applied on the Y channel of two source images. The fused Y channel is implemented by using many fusion rule techniques. The Cb and Cr channels of the source images are fused using principal component analysis (PCA). The proposed method performance is evaluated in terms of PSNR, RMSE and SSIM. The results show that the fusion quality of the proposed algorithm is better than obtained by several other fusion methods, including SWT, PCA with RGB source images and PCA with YCbCr source images.

A Review on Diversified Mechanisms for Multi Focus Image Fusion

In today’s world, the demand of Image Fusion has been drastically inflated owing to limitation of improper image capturing, lack of clarity and quality of various scenes. Image fusion is the process of combining two or more number of images to produce a high-resolution multi spectral image because it maximizes the relevant data from input pictures of a scene into one composite image by reducing uncertainty that additionally minimizes redundancy within the output. In this paper we comprehensively survey the existing methods of image fusion techniques for multi focus image fusion such as Averaging method, Differential Evolution algorithm-based technique, Discrete Wavelet Transform based fusion and Quaternion Wavelet Transform based fusion. This comparison of all the techniques serves as reference for researcher in multi focus image fusion and related work for future analysis.

Image Matting Based Multi-Focus Image Fusion With Optimal Cluster Size

Image fusion is used to intensify the quality of images by combining two images of same scene obtained from different techniques. The present work deals with the effective extraction of pixel information from the source images that hold the key to multi focus image fusion. A solely vicinity-based image matting algorithm that relies on the close pixel clusters in the input images and their trimap, is presented in this article. The pixel cluster size, N plays a significant role in deciding the identity of the unknown pixel. The distance between each unknown pixel from foreground and background pixel clusters has been computed based on minimum quasi Euclidean distance. The minimum distance ratio gives the alpha value of each unknown pixel in the image. Finally, the focus regions are blend together to obtain the resultant fused image. On perceiving the results visually and objectively, it is concluded that proposed method works better in extracting the focused pixels and improving fusion quality, compared with other existing fusion methods.

Efficient Multi Focus Image Fusion Technique Optimized Using MOPSO for Surveillance Applications

This article describes how image fusion has taken giant leaps and emerged as a promising field with diverse applications. A fused image provides more information than any of the source images and it is very helpful in surveillance applications. In this article, an efficient multi focus image fusion technique is proposed in cascaded wavelet transform domain using swarm intelligence and spatial frequency (SF). Spatial frequency is used for computing the activity level and consistency verification (CV) based decision map is employed for acquiring the final fused coefficients. Justification for employing SF and CV is also discussed. This technique performs well compared to existing techniques even when the source images are severely blurred. The proposed framework is evaluated using quantitative metrics, such as root mean square error, peak signal to noise ratio, mean absolute error, percentage fit error, structural similarity index, standard deviation, mean gradient, Petrovic metric, SF, feature mutual information and entropy. Experimental outcomes demonstrate that the proposed technique outperforms the state-of-the art techniques, in terms of visual impact as well as objective assessment.

A Review on Various Techniques of Image Fusion for Quality Improvement of Images

Image fusion is a significant topic in perspective processing. Image fusion is a process of mixing the appropriate data from some images into a single image where the resulting merged picture will be more useful and complete than any of the input images. Multi focus Image fusion is procedure of combining information of several imagery of a view and consequently has \everywhere in focus "image. Lifting technique allows faster implementation of wavelet transform. It requires half number of computations as compared to traditional convolution approach.

Multi Focus Image Fusion using Combined Median and Average Filter based Hybrid Stationary Wavelet Transform and Principal Component Analysis

Poor illumination, less background contrast and blurring effects makes the medical, satellite and camera images difficult to visualize. Image fusion plays the vital role to enhance image quality by resolving the above issues and reducing the image quantity. The combination of spatial and spectral technique Discrete Wavelet Transform and Principal Component Analysis (DWT-PCA) decrease processing time and reduce number of dimensions but down sampling causes lack of shift invariance that results in poor quality final fused image. At first this work uses combined median and average filter that eliminates noise in the image which is caused by illumination, camera circuitry and sensor at preprocessing stage. Then, hybrid Stationary Wavelet Transform and Principal Component Analysis (SWT-PCA) technique is implemented to increase output image accuracy by eliminating down sampling and is not influenced by artifacts and blurring effects. Further, it can overcome the trade-off of Heisenberg’s uncertainty principle by improving accuracy in both domains, time (spatial) as well as frequency (spectral). The proposed combined median and average filter with hybrid SWT-PCA algorithm measures quality parameters, such as peak signal to noise ratio (PSNR), mean squared error (MSE) and normalized cross correlation (NCC) and improved results depict the superiority of the algorithm than existing techniques.