Reconstruction of Compressed Hyperspectral Image Using SqueezeNet Coupled Dense Attentional Net

Abstract

This study addresses image denoising alongside the compression and reconstruction of hyperspectral images (HSIs) using deep learning techniques, since the research community is striving to produce effective results to utilize hyperspectral data. Here, the SqueezeNet architecture is trained with a Gaussian noise model to predict and discriminate noisy pixels of HSI to obtain a clean image as output. The denoised image is further processed by the tunable spectral filter (TSF), which is a dual-level prediction filter to produce a compressed image. Subsequently, the compressed image is analyzed through a dense attentional net (DAN) model for reconstruction by reverse dual-level prediction operation. All the proposed mechanisms are employed in Python and evaluated using a Ben-Gurion University-Interdisciplinary Computational Vision Laboratory (BGU-ICVL) dataset. The results of SqueezeNet architecture applied to the dataset produced the denoised output with a Peak Signal to Noise Ratio (PSNR) value of 45.43 dB. The TSF implemented to the denoised images provided compression with a Mean Square Error (MSE) value of 8.334. Subsequently, the DAN model executed and produced reconstructed images with a Structural Similarity Index Measure (SSIM) value of 0.9964 dB. The study proved that each stage of the proposed approach resulted in a quality output, and the developed model is more effective to further utilize the HSI. This model can be well utilized using HSI data for mineral exploration.

Assessing Soil Nutrient Content and Mapping in Tropical Tamil Nadu, India, through Precursors IperSpettrale Della Mission Applicative Hyperspectral Spectroscopy

“…This space-borne mission collects data from three different spectral regions: the very near infrared region (VNIR), the short wave infrared region (SWIR), and the panchromatic region (PR). The VNIR spectrum spans 400 nm to 1010 nm, with a spectral resolution of 12 nm, 66 continuous spectrum bands, and a spatial resolution of 30 m [32,33]. In SWIR, the spectrum ranges from 920 nm to 2505 nm with 12 nm spectral resolution, 171 endless spectrum bands, and 30 m spatial resolution.…”

Camera and LiDAR Fusion for Urban Scene Reconstruction and Novel View Synthesis via Voxel-Based Neural Radiance Fields

“…Given the camera viewpoint P and loss function in Equation ( 8), we calculate the rendered depth of the ray set R (P) with existing LiDAR point cloud data from this viewpoint. The sparse point cloud depth loss function minimizes the difference between this rendered depth and the actual depth D gt (r): (10) To enhance the structural constraints of the 3D model, this paper introduces the slanted plane model into the NeRF training framework. It assumes that all pixels within a superpixel are located on the same 3D plane.…”

“…3D reconstruction [5,6] can be used not only for data augmentation but also for direct 3D modeling of urban scenes [7]. Specifically, in the remote sensing mapping [8][9][10][11][12], it can generate high-precision digital surface models using multi-view satellite images [13,14] and combine the diversity of virtual environments with the richness of the real-world, generating more controllable and realistic data than simulation data.…”