Research on the Application of Deep Learning in Food Detection

DOI: https://doi.org/10.62517/jike.202604408

Author(s)

Zheng Sen*, Orgil Demberel

Affiliation(s)

School of Applied Sciences, Mongolian University of Life Sciences, Ulaanbaatar, Mongolia

Abstract

With the continuous development of deep learning-related technologies, image recognition has gradually shifted from traditional manual feature methods to data-driven representation learning models. In the context of dish detection, due to the large number of categories, small visual differences between categories, and complex background environments, the image recognition process poses certain challenges. This paper combines typical models such as convolutional neural networks, deep belief networks, and recurrent neural networks to analyze their feature learning mechanisms in fine-grained image recognition. The research shows that deep learning can enhance recognition capabilities in complex visual scenes, but model performance is still affected by factors such as data quality and training strategies, and there is still room for further optimization of related methods.

Keywords

Deep Learning; Image Recognition; Dish Detection; Convolutional Neural Network; Representation Learning

References

[1] LeCun Y, Bengio Y, Hinton G. Deep learning[J]. nature, 2015, 521(7553): 436-444. [2] Krizhevsky A, Sutskever I, Hinton G E. ImageNet classification with deep convolutional neural networks[J]. Advances in neural information processing systems, 2012, 25. [3] Goodfellow I, Bengio Y, Courville A, et al. Deep learning[M]. Cambridge: MIT press, 2016. [4] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[J]. arXiv preprint arXiv:1409.1556, 2014. [5] He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2016: 770-778. [6] Ren S, He K, Girshick R, et al. Faster R-CNN: Towards real-time object detection with region proposal networks[J]. IEEE transactions on pattern analysis and machine intelligence, 2016, 39(6): 1137-1149.