PLANT STRESS DETECTION USING THE LIMB OF THE LEAF BY MULTI SPECTRAL IMAGERY Mama SANGARE, BA Abdramane, Zoueu T. Jeremie Laboratory of Optical Spectroscopy and Sciences of Atmosphere (LOSSA) Department of physics - Faculty of Sciences and Techniques University of Sciences, Techniques and Technologies of Bamako (MALI) Reflex - Percentage error : 0

Introduction

Measures

A leaf is made of limb, secondary and principal vein. But the photosynthetic radiation occurs in the limb part of the leaf. Studies undertaken on the limb showed that it is composed of water and many mineral salts such as calcium, potassium, sodium, magnesium, phosphorus. For a leaf, a deficiency of one of these essential elements can cause a stress leading to a reduction of its growth. We propose in this work a study on the hydric stress by using a multimode multispectral microscopy. The goal is to show the capacities of our microscopy to detect the various stresses and diseases of some tropical plants starting from the spectral data of the microscopic components of the limb of the leaf.

We selected some leaves of plants to knowing: leaf of corn, mango tree, cassava and pawpaw tree. We created for each one the stress conditions (figure 1) and measurements were taken with the multispectral microscopy on the limb of each leaf of plants.

10

Numbe r wrong classified object: 0

Fresh leaf - Maize Hot leaf - Maize Dry leaf - Maize

PC3

5

0

-5

-10 -6

-4

-2

0

2

4

6

8

PC1

Figure 3: leaf corn stress projection using PCA and LDA in reflection mode. Scatt - Percentage error : 0

15 10

Number wrong classifie d object : 0

Fresh leaf - maize Hot leaf- Maiz e Dry leaf - Maiz e

PC2

5 0 -5

Methods -10 -4

-2

0

2

4

6

8

10

PC1

Figure 4: leaf corn stress projection using PCA and LDA in diffusion mode.

The algorithm used for this study can be described according to this diagram:

Conclusion Figure1: RGB images in the three geometries for the corn leaf with different stress

Results the results obtained after using PCA and LDA show us that the various created stress have been well discriminated (figure 3,4,5) Trans - Percentage error : 1.6667

Number wrong classified object : 1

4 Fresh leaf - Maize Hot leaf - Maize Dry leaf- Maize

3

References

2 PC3

The goal of this work concerning the detection of the various types of stress of leaf of plants was reached. Thanks to the techniques and steps adopted in this work, the conditions of stress created for each leaf of plant were characterized and discriminates. The results shows us that our spectral imagor can be used in thorough studies for the detection and discrimination of potential diseases LAURENT BOUSQUET. 2007; ’’ Mesure et modélisation des propriétés optiques spectrales et directionnelles des feuilles’’; Télédétection de l’environnement; Doctorat 3ème cycle; P. 5 -12.

1 0 -1 -2 -3

-6

-4

-2

0

2

4

6

GREGORY A. CARTER, 1994; ‘’ Ratios of leaf reflectances in narrow wavebands as indicators of plant stress’’; INT. J. Remote Sensing; Vol. 15; P. 697 – 699.

PC2

Figure 2: leaf corn stress projection using PCA and LDA in transmission mode.

Al-ABBAS ET AL 1974; ‘’ Spectra of normal and nutrient – deficient maize leaves ‘’; P. 16.