Paper Contents
Abstract
The paper outlines various techniques for identifying nutrient deficiencies that can be used in a variety of settings and help to increase food safety. The recommended method or measure starts by dividing the image of the leaf that is given as input into large pieces. Next, each pixel block of the leaf is merged with a series of convolutional neural networks. These CNNs are mainly responsible for identifying the lack of nutrient deficiency from each block. Each and every individual CNN is trained especially for a particular kind of nutrient deficit. Then the outcomes of all the corresponding CNNs are then merged altogether using the winner takes all strategy to obtain a single reply for each block. At last, all the responses are converted into a single response to provide a final response for the entire leaf using a multi-layer perceptron. The suggested idea is tested on a set of black gramme plant leaves cultivated in various areas. A total of five nutrient deficient leaves including calcium, iron, potassium, magnesium and nitrogen and a complete nutrition class is proposed in this method.
Copyright
Copyright © 2023 Bidawat Tarajit Singh. This is an open access article distributed under the Creative Commons Attribution License.
