Mulberry leaves provide a lot of medicinal and industrial uses along with being an excellent food source for silkworms. But one has to know the correct type of mulberry leaf to make use of it correctly. This research paper aims to achieve a high percentage of accuracy in mulberry leaf image classification using deep learning models. The dataset chosen is a mulberry leaf dataset consisting of 10 different classes. The reason behind using deep learning is the efficiency of such neural network models which can easily simulate the decision-making power of a human brain. A custom model has been developed for the research which uses convolution neural networks to achieve an accuracy of over 92.25%. This study also compares the performance of this custom model with other pre-trained models which are deemed to be highly accurate for the work of image classification and are frequently used in developing high-end models for a bigger use.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

DeepLeaf: A Custom CNN Approach for Mulberry Leaf Classification

  • Tripti Mishra,
  • Vanshaj Singhal,
  • Yashaswat Verma,
  • Monika,
  • Manish Raj

摘要

Mulberry leaves provide a lot of medicinal and industrial uses along with being an excellent food source for silkworms. But one has to know the correct type of mulberry leaf to make use of it correctly. This research paper aims to achieve a high percentage of accuracy in mulberry leaf image classification using deep learning models. The dataset chosen is a mulberry leaf dataset consisting of 10 different classes. The reason behind using deep learning is the efficiency of such neural network models which can easily simulate the decision-making power of a human brain. A custom model has been developed for the research which uses convolution neural networks to achieve an accuracy of over 92.25%. This study also compares the performance of this custom model with other pre-trained models which are deemed to be highly accurate for the work of image classification and are frequently used in developing high-end models for a bigger use.