Super Resolution Reconstruction Based on Adaptive Detail Enhancement for Zy-3 Satellite Images

Abstract

ABSTRACT:Super-resolution reconstruction of sequence remote sensing image is a technology which handles multiple low-resolution satellite remote sensing images with complementary information and obtains one or more high resolution images. The cores of the technology are high precision matching between images and high detail information extraction and fusion. In this paper puts forward a new image super resolution model frame which can adaptive multi-scale enhance the details of reconstructed image. First, the sequence images were decomposed into a detail layer containing the detail information and a smooth layer containing the large scale edge information by bilateral filter. Then, a texture detail enhancement function was constructed to promote the magnitude of the medium and small details. Next, the non-redundant information of the super reconstruction was obtained by differential processing of the detail layer, and the initial super resolution construction result was achieved by interpolating fusion of non-redundant information and the smooth layer. At last, the final reconstruction image was acquired by executing a local optimization model on the initial constructed image. Experiments on ZY-3 satellite images of same phase and different phase show that the proposed method can both improve the information entropy and the image details evaluation standard comparing with the interpolation method, traditional TV algorithm and MAP algorithm, which indicate that our method can obviously highlight image details and contains more ground texture information. A large number of experiment results reveal that the proposed method is robust and universal for different kinds of ZY-3 satellite images.

Blind turbulent image deblurring through dual patch-wise pixels prior

“…53 can obtain the restoration result with a better visual effect. For real images wherein the ground-truth images are not available, entropy, 54 , 55 blind/referenceless image spatial quality evaluator (BRISQUE), 56 and blind multiple pseudo reference images (BMPRI) 57 are used to evaluate the quality of the restored images. Entropy is an important measure of image information index and is ideal for evaluating the quality of satellite images.…”

Deep Learning for Multiple-Image Super-Resolution

“…Among other methods, adaptive Wiener filter [9] and random Markov fields [16] were used to specify the IM. Zhu et al proposed adaptive detail enhancement (SR-ADE) [20] for reconstructing satellite images-a bilateral filter arXiv:1903.00440v1 [cs.CV] 1 Mar 2019 is employed to decompose the input images and amplify the high-frequency detail information.…”

“…For all these metrics, higher values indicate higher similarity between the reconstruction outcome and the reference image. EvoNet is compared with two single-image SRR methods: SRR based on wavelet transform (SR-DWT) [4] and ResNet [15], and with three multiple-image ones: GPA [19], SR-ADE [20], and EvoIM [11]. EvoIM (also exploited in EvoNet) was trained separately for artificially-degraded images and for real satellite data, as reported in [11], using PSNR HF [2] as the fitness function (there were no overlaps between training and test sets).…”

Single satellite imagery simultaneous super-resolution and colorization using multi-task deep neural networks

“…Over the past five years, a considerable number of image SR works have addressed to reconstruct HR satellite imagery from LR inputs. Usually, these methods are divided into two categories: multiple images reconstruction (Pickup, 2007;Zhang et al, 2014;Hung et al, 2016;Zhu et al, 2016;Brodu, 2016;Alvarez-Ramos et al, 2016) or single image SR (Liebel and Körner, 2016;Patrick, 2016).…”

“…Due to possible resolution differences in multi-angle remote sensing images over the same scene, adaptive weighting schemes (Zhang et al, 2014) are utilized to reconstruct HR satellite imagery. In addition, adaptive multi-scale detail enhancement measures (Zhu et al, 2016) were attempted for multiple LR satellite images SR. Moreover, sparse representation (Alvarez-Ramos et al, 2016) has been employed to deal with overlapping blocks for satellite image SR.…”