Dark Subtraction Research Articles

The reptile optimized deep learning model for land cover classification of the uppal earth region in telangana state using satellite image fusion

This study addresses challenges in land use and cover identification using remote sensing (RS) imagery, focusing on the Uppal region. By leveraging deep learning models, particularly an optimized ResNext-50 architecture, we aim to enhance efficiency and accuracy in classifying land features. Our approach integrates Landsat-8 and hyper-spectral satellite data, utilizing preprocessing techniques like dark subtraction, stacking, merging, and spectral enhancement. Principal Component Analysis (PCA) is applied to streamline high-dimensional feature sets obtained from pre-processed spectral data. We further employ hybrid NSCT-FDCT fusion for integrating Landsat-8 and hyperspectral images. The resulting fused image is fed into our classification process, utilizing the modified ResNext50 (Deep Learning Architecture) model with Reptile Search Optimization for weight link optimization. Notably, our proposed method achieves impressive outcomes: 97% accuracy, 96% sensitivity, 99% specificity, 3% error, 97% precision, and a 95% Matthew Correlation Coefficient. This demonstrates the efficacy of our approach in predicting diverse land covers within the Uppal region, showcasing the potential of Landsat-8 and Hyper-spectral data for accurate land use and cover identification.

Investigation into the Affect of Chemometrics and Spectral Data Preprocessing Approaches upon Laser-Induced Breakdown Spectroscopy Quantification Accuracy Based on MarSCoDe Laboratory Model and MarSDEEP Equipment

As part of China’s Tianwen-1 Mars mission, the Mars Surface Composition Detector (MarSCoDe) instrument on the Zhurong rover adopts laser-induced breakdown spectroscopy (LIBS) to perform chemical component detection of the materials on the Martian surface. However, it has always been a challenging issue to achieve high accuracy in LIBS quantification. This study investigated the effect of chemometrics and spectral data preprocessing approaches on LIBS quantification accuracy based on different chemometrics algorithms and diverse preprocessing methods. A total of 2340 LIBS spectra were collected from 39 kinds of geochemical samples by a laboratory duplicate model of the MarSCoDe instrument. The samples and the MarSCoDe laboratory model were placed in a simulated Martian atmosphere environment based on equipment called the Mars-Simulated Detection Environment Experiment Platform (MarSDEEP). To quantify the concentration of MgO in the samples, we employed two common LIBS chemometrics; i.e., partial least squares (PLS) and a back-propagation neural network (BPNN). Meanwhile, in addition to necessary routine preprocessing such as dark subtraction, we used five specific preprocessing approaches, namely intensity normalization, baseline removal, Mg-peak wavelength correction, Mg-peak feature engineering, and concentration range reduction. The results indicated that the performance of the BPNN was better than that of the PLS and that the preprocessing of Mg-peak wavelength correction had the most prominent effect to improve the quantification accuracy. The results of this study are expected to provide inspiration for the processing and analysis of the in situ LIBS data acquired by MarSCoDe on Mars.

A radiometric calibration for the Clementine HIRES camera

We have successfully characterized the Clementine HIRES multispectral sensor entirely from inflight measurements. Relative calibration is obtained through dark subtraction (additive) and sensor spatial nonuniformity (multiplicative) on a pixel‐by‐pixel basis. Relative residuals from frame‐to‐frame are typically less than 1%. Absolute coefficients are obtained through comparisons with calibrated Clementine UVVIS images acquired nearly simultaneously with HIRES sequences. Residuals between the coregistered UVVIS and HIRES calibrated images are typically 5% but may be as high as 20%. The residual is found most likely to be the result of increased scattered light in the UVVIS relative to the HIRES. Application of the calibration allows for seamless mosaicking of monochromatic and color sequences aiding in interpretation of surface features of the Moon at the scale of about 25 m/pixel. The calibrated HIRES color sequences represent the highest resolution spectral image data of the Moon.