Fruit Image Classification Using Convolution Neural Networks
DOI:
https://doi.org/10.54097/hset.v34i.5430Keywords:
deep learning, fruit classification, convolutional neural network, image classification.Abstract
Artificial intelligence has been used in many places in people's daily life, and there are more and more methods to classify objects by using deep learning. However, at present, the technique of fruit classification is mainly manual classification, and the accuracy of mechanical classification needs to be improved. The ideas for fruit recognition are primarily focused on distinguishing the shape, texture, and colour of fruits. Therefore, this paper uses deep learning technology since the performance of deep learning in the field of computer vision is better than that of traditional machine learning. The convolutional neural network (CNN) of deep learning will automatically learn the features of different fruit images to establish a model for predicting fruit types. In this paper, four CNNs models with different structures are compared with the prediction results for 131 different types of fruit. The data showed that the best model provided 98.2% accuracy.
Downloads
References
World Health Organization. (2005). Fruit and vegetables for health: report of the Joint FAO.
Yang, J., Qiao, P., Li, Y., & Wang, N. (2019). Machine learning classification problems and algorithms, 35(6), 36-40.
Chaganti, S. Y., Nanda, I., Pandi, K. R., Prudhvith, T. G., & Kumar, N. (2020, March). Image Classification using SVM and CNN. In 2020 International Conference on Computer Science, Engineering and Applications (ICCSEA) (pp. 1-5). IEEE.
Baranwal, S. K., Jaiswal, K., Vaibhav, K., Kumar, A., &Srikantaswamy, R. (2020, July). Performance analysis of brain tumour image classification using CNN and SVM. In 2020 Second International Conference on Inventive Research in Computing Applications (ICIRCA) (pp. 537-542). IEEE.
Wang, P., Fan, E., & Wang, P. (2021). Comparative analysis of image classification algorithms based on traditional machine learning and deep learning. Pattern Recognition Letters, 141, 61-67.
Zhang, C. (2013). The challenges of machine learning. Science China Press: Information Science, 43(12), 1612-1623.
Xu, B., Cen, K., Huang, J., Shen, H., & Cheng, X. (2020). A Survey on Graph Convolutional Neural Network. Chinese Journal of Computers, 43(5), 755-780.
Sarıgül, M., Ozyildirim, B. M., &Avci, M. (2019). Differential convolutional neural network. Neural Networks, 116, 279-287.
Keras (n.d.). Keras documentation: About keras. Keras. From https://keras.io/about/
Oltean, M. (2021, September 12). Fruits 360. Kaggle. From https://www.kaggle.com/datasets/moltean/fruits
Ge, Y., Liu, H., Wang, Y., Xu, B., Zhou, Q., & Shen, F. (2022). Survey on Deep Learning Image Recognition in Dilemma of Small Samples. Journal of Software, 33(1), 192-210.
Mairal, J., Koniusz, P., Harchaoui, Z., & Schmid, C. (2014). Convolutional kernel networks. Advances in neural information processing systems, 27.
Liu, W., Liang, X., & Qu, H. (2016). Learning performance of convolutional neural networks with different pooling models. Journal of Image and Graphics, 21(9), 1178-1190.
Xie, T. (2021). Research of dropout method based on convolutional neural network. University of Electronic Science and Technology of China.
Lv, F., & Liu, T. (2021). A deep learning algorithm for image recognition with self-enhancement of learning rate. Computer Application and Software, 10.3969.
Deng, J., Zhang, S., Zhang, J., Xun, Y., & Liu, A. (2020). Loss function and application research in supervised learning. Big Data Research, 10.11959.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
How to Cite
