Study Of Microlens-based Light Field Camera All-in-focus Image Fusion And Depth Map Calculation

With the development of Computational Photography and light field imaging theory,light field became a hot topic during the past ten years in many research fields.Compared with the traditional camera,a micro-lens array was added after the main-lens,which made the light field camera can record the location and direction information of the light.Light field camera has more advantages in image post-processing and applications.The most important technical high light of light field camera is all the refocused image can be calculated at any depth after a single exposure.Based on this feature,many high-quality texture image acquisition methods and high-accuracy depth map calculation techniques are deeply studied.Not limited by the focusing system in multi-focused image capture of traditional camera,the all-in-focusing image fusion and depth estimation based on digital refocusing became the main research fields in light field camera,which will exert a great impact on the texture and depth map super-resolution and light-field video generation.All-in-focusing image and depth map calculation based on micro-lens light field camera are studied in this paper.The main research work and high lights are following:1.A light field all-in-focus image fusion method based on wavelet transform is proposed.Compared with the spatial image fusion method,the proposed method can effectively avoid the block artifacts and obtain a fused image with high quality.Firstly,the refocused images used for the all-in-focus image calculation can be computed through shearing and projecting the 4D light field captured by the microlens-based light field camera.Then the Wavelet transform are applied to the refocused images and the high frequency and low frequency sub-images are extracted respectively.Finally the balanced Laplace operator and pixel visibility function are proposed to evaluate the sharpness of the sub-image to get a high quality fusion image.Experiment results prove the validity of the proposed method.The raw images captured by Lytro light field camera are used to calculate the all-in-focus image.Compared with the traditional Wavelet transform-based image fusion methods,the visual effect is better and the quantitative indices are also improved with the proposed method.2.An optimized depth map estimation algorithm based on balanced Laplace operator is proposed,which can reconstruct the depth with higher accuracy when the shootingenvironment with complex depth information.The offset of the second derivative with different sign in x and y direction in the Laplace operator and the insufficient of direction information of the second derivative during the depth estimation are considered in the proposed algorithm.An improved Laplace operator based on region is designed,which can improve the energy balancing in Laplace operator through adding the numbers and directions of the second derivative.The effectiveness of the proposed method is demonstrate by experiment results.Our algorithm can calculate a more reasonable depth map with more details when the shooting environment is complex...