Estimation of heavy metal concentrations (Cd and Pb) in plant leaves using optimal spectral indicators and artificial neural networks.

The necessity of continuously monitoring the agricultural products in terms of their health has enforced the development of rapid, low-cost, and non-destructive monitoring solutions. Heavy metal contamination of the plants is known as a source of health threats that are made by their proximities with pollutant soil, water, and air. In this paper, a method was proposed to measure lead (Pb) and cadmium (Cd) contamination of plant leaves through field spectrometry as a low-cost solution for continuous monitoring. The study area was Mahneshan county of Zanjan province in Iran with rich heavy metal mines that have more potential for plant contamination. At first, we collected different plant samples throughout the study area and measured the Pb and Cd concentrations using ICP-AES, in which we observed that the concentrations of Pb and Cd are in the range of 1.4 ~ 282.6 and 0.3 ~ 66.7 µgg-1, respectively, and then we tried to find the optimum estimator model through a multi-objective version of genetic algorithm (GA) optimization that finds simultaneously the structure of an artificial neural network and its input features. The features extracted from the raw spectrums have been collimated to be compatible with the Sentinel-2 multispectral bands for the possibility of further developments. The results demonstrate the efficiency of the optimum estimator model in estimation of the leaves' Pb and Cd contamination, irrespective of the plant type, which has reached the R2 of 0.99 and 0.85 for Pb and Cd, respectively. Additionally, the results suggested that the 783-, 842-, and 865-nm spectral bands, which are similar to the 7, 8, and 8a sentinel-2 spectral bands, are more efficient for this purpose.

Epipolar Resampling of Cross-Track Pushbroom Satellite Imagery Using the Rigorous Sensor Model.

Epipolar resampling aims to eliminate the vertical parallax of stereo images. Due to the dynamic nature of the exterior orientation parameters of linear pushbroom satellite imagery and the complexity of reconstructing the epipolar geometry using rigorous sensor models, so far, no epipolar resampling approach has been proposed based on these models. In this paper for the first time it is shown that the orientation of the instantaneous baseline (IB) of conjugate image points (CIPs) in the linear pushbroom satellite imagery can be modeled with high precision in terms of the rows- and the columns-number of CIPs. Taking advantage of this feature, a novel approach is then presented for epipolar resampling of cross-track linear pushbroom satellite imagery. The proposed method is based on the rigorous sensor model. As the instantaneous position of sensors remains fixed, the digital elevation model of the area of interest is not required in the resampling process. Experimental results obtained from two pairs of SPOT and one pair of RapidEye stereo imagery with different terrain conditions shows that the proposed epipolar resampling approach benefits from a superior accuracy, as the remained vertical parallaxes of all CIPs in the normalized images are close to zero.