| Autor: |
P., Josephin Shermila, V., Seethalakshmi, A., Ahilan, M., Devaki |
| Předmět: |
|
| Zdroj: |
International Journal of Intelligent Engineering & Systems; 2024, Vol. 17 Issue 4, p289-298, 10p |
| Abstrakt: |
Fruits are structures that contain seeds that are created from the ovaries of only blooming plants. A fruit is a soft, component of a blooming plant that bears seeds. It is created from angiosperm ovaries and is unique to this plant group. In this paper proposed a novel Deep Learning-based Fruits Items Classification (SOOTY FRUIT) method to classify the variety of Fruits items from the dataset with their calorie values (CV). Initially, the images are processed using the Gaussian adaptive bilateral (GAB) Filter approach to improve image quality and eliminate noise. Consequently, the segmented images are pre-processed utilizing Attention V-net algorithm. The extracted features are then normalized utilized the Sooty Tern Optimization Algorithm (STOA). Fruits items are classified using Ghostnet based on these relevant features. As compared to existing methods, the proposed SOOTY FRUIT shows better results in terms of Accuracy. The accuracy of the proposed technique can be as high as 99.95%, while that of traditional models like the attention-based densely connected convolutional networks with convolution autoencoder (CAE-AND), Faster-Region-based Convolutional Neural Network (F-RCNN), and Attention Fusion Network (AFN) is 84.9%, 87.58%, and 93.91%, respectively. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
