Published December 7, 2023 | Version v1
Software Open

Classification of nutrition state in maize leaves by transfer learning

Creators

  • 1. ROR icon Universidad EAFIT

Description

Monitoring the nutritional status of crops is crucial in agricultural management. Given that nutritional deficiencies primarily manifest through visual characteristics, artificial vision stands out as a competitive choice to assess the nutritional status of individual plants. However, to train a supervised artificial vision system driven by convolutional neural networks (CNNs), a high amount of data, properly formatted and labeled is necessary. That's why transfer learning techniques address some of these challenges.

Table of contents

To address the timely identification of nutritional disorders, it is proposed a bedrock composed of: 

Load and clean datastore.rar

  • Preprocessing of data: Code for resizing and cropping images to increase the number of original images fivefold.
  • Split the data: MATLAB code to split data into training, validation, and test sets.

MATLAB_networks.rar

  • ImageDataStores: Containing training, validation, and test sets.
  • Transfer learning: MATLAB code to run five transfer learning models <net>= VGG16, ResNet50, GoogLeNet, DenseNet201, and MobileNetV2 on the imagesets.

Output: 

The output of each network script are:

  • Traininfo_<net>.csv: Having Training_accuracy, Validation_accuracy, Training_loss, Validation_loss info
  • OutPred_<net>.csv: Predicted labels
  • Outtrue_<net>.csv: True labels
  • confussionMatrix__<net>.pdf: confusion matrix

Files

Files (44.7 kB)

Name Size Download all
md5:22d866d8990eb6e57b9712275701f427
8.3 kB Download
md5:1ee925afbfabf4e1834a624cfe3a29cc
36.4 kB Download