Polarimetric observables for the enhanced visualization of plant diseases

Abstract

This paper highlights the potential of using polarimetric methods for the inspection of plant diseased tissues. We show how depolarizing observables are a suitable tool for the accurate discrimination between healthy and diseased tissues due to the pathogen infection of plant samples. The analysis is conducted on a set of different plant specimens showing various disease symptoms and infection stages. By means of a complete image Mueller polarimeter, we measure the experimental Mueller matrices of the samples, from which we calculate a set of metrics analyzing the depolarization content of the inspected leaves. From calculated metrics, we demonstrate, in a qualitative and quantitative way, how depolarizing information of vegetal tissues leads to the enhancement of image contrast between healthy and diseased tissues, as well as to the revelation of wounded regions which cannot be detected by means of regular visual inspections. Moreover, we also propose a pseudo-colored image method, based on the depolarizing metrics, capable to further enhance the visual image contrast between healthy and diseased regions in plants. The ability of proposed methods to characterize plant diseases (even at early stages of infection) may be of interest for preventing yield losses due to different plant pathogens.

Automatic pseudo-coloring approaches to improve visual perception and contrast in polarimetric images of biological tissues

“…In both methods, the construction of the corresponding pseudo-coloring functions is based on the selection of a triplet of polarization-based figures or observables. As previously mentioned, we selected depolarization-related observables because as reported in literature 16 , 17 , 28 – 30 , and based in our previous experience we know that they are suitable to characterize biological samples 1 , 3 – 5 . To reinforce this argument, in section 1 of Supplemental document, we show the images of different biological structures related to a representative collection of different polarimetric observables used in the literature, as well as the images corresponding to IPPs and CPs, to show how the latter spaces give rise to greater visual contrast.…”

“…In this work, we have provided optimized pseudo-colored functions based on polarimetric spaces, which enhance other approaches we previously presented in literature 1 , 24 , 25 , 30 . However, we want to note that the application of pseudo-colored functions is not limited to polarimetric spaces 33 – 36 , and there exist other approaches, as image segmentation and coloring approaches 37 – 40 , that lead to interesting results in terms of structures visualization.…”

Optimizing the classification of biological tissues using machine learning models based on polarized data

“…These four algorithms are described in Section 3 of Supporting information document. This led to 12 final models: (1) ANN with A, (2) ANN with B, (3) ANN with C, (4) SVM with A, (5) SVM with B, (6) SVM with C, (7) LGBM with A, (8) LGBM with B, (9) LGBM with C, (10) XGBoost with A, (11) XGBoost with B, and (12) XGBoost with C.…”

“…The use of polarization‐based techniques for the study of animal [1, 2], human tissues [3–8] and vegetal samples [9–11] is a well‐stablished field of work which is in constant development [12–14]. Polarimetry comprises a wide range of applications including biomedicine [2], materials characterization [15], atmospheric pollution [16, 17], plant science [10], etc.…”

“…For instance, the polarimetric microscope [10] allowed the visualization of nuclear complexes in cell division processes [25]. Furthermore, it has been demonstrated the capability of the polarimetric observables to enhance the image contrast between healthy and diseased tissues, as well as the revelation of wounded regions, in plant specimens [9,11].…”