Neural network segmentation of images from stained cucurbits leaves with colour symptoms of biotic and abiotic stresses

Abstract

The increased production of Reactive Oxygen Species (ROS) in plant leaf tissues is a hallmark of a plant's reaction to various environmental stresses. This paper describes an automatic segmentation method for scanned images of cucurbits leaves stained to visualise ROS accumulation sites featured by specific colour hues and intensities. The leaves placed separately in the scanner view field on a colour background are extracted by thresholding in the RGB colour space, then cleaned from petioles to obtain a leaf blade mask. The second stage of the method consists in the classification of within mask pixels in a hue-saturation plane using two classes, determined by leaf regions with and without colour products of the ROS reaction. At this stage a two-layer, hybrid artificial neural network is applied with the first layer as a self-organising Kohonen type network and a linear perceptron output layer (counter propagation network type). The WTA-based, fast competitive learning of the first layer was improved to increase clustering reliability. Widrow-Hoff supervised training used at the output layer utilises manually labelled patterns prepared from training images. The generalisation ability of the network model has been verified by K-fold cross-validation. The method significantly accelerates the measurement of leaf regions containing the ROS reaction colour products and improves measurement accuracy.

Automated image analysis for quantification of histochemical detection of reactive oxygen species and necrotic infection symptoms in plant leaves

“…Then the first layer pixel subclasses can be merged interactively in a dedicated dialog window to achieve final classification of regions stained with DAB and NBT. The algorithm ensured classification with high accuracy as shown by the leaf pixel classification error of 1.42% (Gocławski et al 2012). The algorithm execution is independent of specific color of leaf blade regions.…”

“…The quantification of ROS production given as a percentage of stained leaf area was achieved by counting the leaf image pixels with known colors corresponding to the respective reaction products. The images of stained leaves were scanned to JPEG files using a standard flatbed scanner with 200/300 dpi and 24 bit/pixel resolutions and then were analyzed by an algorithm run in MATLAB environment on a Windows computer (Gocławski et al 2012). Briefly, each leaf was placed on a scanner glass in a fixed direction, with the background color differing from the colors found inside of a leaf blade, namely blue for DAB and red for NBT staining.…”

Prediction of corrections for the Polish time scale UTC(PL) using artificial neural networks

“…Moreover, special types of neural networks are used for adaptive synthesis of the wavelet transform [8], image segmentations [9] and systems identification and diagnosis [10,11], which proves that ANNs can be used in many technical areas. Based on this fact, for the prediction of corrections, the method based on artificial neural networks is examined in detail in this paper.…”

A differential evolution approach to dimensionality reduction for classification needs

“…Methods used for classification vary from Bayes classifiers and k-NN algorithms , Artificial Neural Networks (ANNs) (Debska and Guzowska-Swider, 2011;Gocławski et al, 2012) and Support Vector Machines (SVMs) (Hsu and Lin, 2002;Jeleń et al, 2008) to classifier ensembles (Woźniak and Krawczyk, 2012), The classification accuracy depends greatly on the method used, but also on the underlying problem, i.e., the characteristics of the data on which the classification method is applied.…”