ABSTRACT
Near-infrared (NIR) spectroscopy has been widely used to determine the varieties and chemical properties of agricultural and food products. The major advantage of NIR spectroscopy is that the analysis is carried out in a simple, fast, and non-destructive manner, making it suitable for food applications. As the first step in applying NIR spectroscopy for fruit recognition and analysis in Vietnam, this paper presents deep neural networks (DNNs) based solutions for automatic recognition of several kinds of fruits. We compared two proposed DNN architectures based on Convolutional Neural Network (CNN) and Residual Network (ResNet). Additionally, we proposed feature extraction methods using the first and second derivatives of the Savitzky-Golay (SG) filtered normalized NIR data. Experimental results show that the deep learning approach combined with reasonable feature extraction process can achive the accuracy of approximately 99% for the task of classifying five types of fruits including Apple, Avocado, Dragon Fruit, Guava, and Mango. The ResNet-based model is more compact and has slightly better recognition performance than the CNN-based one. The inclusion of the first and second derivatives of SG-smoothed normalized spectra improves the recognition accuracy of the proposed DNN models by more than 8%. Moreover, the recognition performance of the proposed DNN models surpasses that of traditional classifiers, including k-nearest neighbors, Naive Bayes, and support vector machine. Our proposed methods were proved to be robust against the freshness of fruits, the NIR device's calibration parameters, and the measurement position on the body of fruits.
- Jan U.Porep, Dietmar R. Kammerer, and Reinhold Carlea. 2015. On-line application of near infrared (NIR) spectroscopy in food production. Trends in Food Science & Technology 46, 2 (December 2015), 211–230. https://doi.org/10.1016/j.tifs.2015.10.002Google ScholarCross Ref
- Jia-Huan Qu, Dan Liu, Jun-Hu Cheng, Da-Wen Sun, Ji Ma, Hongbin Pu, and Xin-An Zeng. 2015. Applications of near-infrared spectroscopy in food safety evaluation and control: a review of recent research advances. Critical Reviews in Food Science and Nutrition, 55, 13 (May 2015), 1939–1954. https://doi.org/10.1080/10408398.2013.871693Google ScholarCross Ref
- Yongni Shao, Yidan Bao, and Yong He. 2011. Visible/near-infrared spectra for linear and nonlinear calibrations: A case to predict soluble solids contents and pH value in peach. Food and Bioprocess Technology 4 (November 2011), 1376–1383. https://doi.org/10.1007/s11947-009-0227-6Google Scholar
- Wenchuan Guo, Lijie Fang, Dayang Liu, and Zhuanwei Wang. 2015. Determination of soluble solids content and firmness of pears during ripening by using dielectric spectroscopy. Computers and Electronics in Agriculture 117 (September 2015), 226–233. https://doi.org/10.1016/j.compag.2015.08.012Google ScholarDigital Library
- Jiangbo Li, Wenqian Huang, Chunjiang Zhao, Baohua Zhang. 2013. A comparative study for the quantitative determination of soluble solids content, pH and firmness of pears by Vis/NIR spectroscopy. Journal of Food Engineering 116, 2 (May 2013), 324–332. https://doi.org/10.1016/j.jfoodeng.2012.11.007Google ScholarCross Ref
- Paloma A. M. Nascimento, Lívia C. de Carvalho, Luis C. C. Júnior, Fabíola M. V. Pereira, Gustavo H. de Almeida Teixeira. 2016. Robust PLS models for soluble solids content and firmness determination in low chilling peach using near-infrared spectroscopy (NIR). Postharvest Biology and Technology 111 (January 2016), 345–351. https://doi.org/10.1016/j.postharvbio.2015.08.006Google Scholar
- Peano, C., Reita, G. and Chiabrando, V., 2006. Firmness and soluble solids assessment of nectarines by NIRS spectroscopy. ISHS Acta Horticulturae 713 (July 2006), 465–469. https://doi.org/10.17660/ActaHortic.2006.713.70Google Scholar
- Lingling Li; Yuan Wu, Lian Li, and Bingqing Huang. 2017. Rapid detecting SSC and TAC of peaches based on NIR spectroscopy. In 2017 2nd IEEE International Conference on Computational Intelligence and Applications (ICCIA). IEEE, Beijing, China, 312–317. DOI: 10.1109/CIAPP.2017.8167229Google Scholar
- Panmanas Sirisomboon, MunehiroTanaka, Takayuki Kojima, and Phil Williams. 2012. Nondestructive estimation of maturity and textural properties on tomato ‘Momotaro'by near infrared spectroscopy. Journal of Food Engineering 112, 3 (October 2012), 218–226. https://doi.org/10.1016/j.jfoodeng.2012.04.007Google ScholarCross Ref
- Frédéric Kosmowski and Tigist Worku. 2018. Evaluation of a miniaturized NIR spectrometer for cultivar identification: The case of barley, chickpea and sorghum in Ethiopia. PLoS ONE 13, 3 (March 2018). https://doi.org/10.1371/journal.pone.0193620Google ScholarCross Ref
- Alex Krizhevsky, Ilya Sutskever, and Geoffrey E. Hinton. 2012. ImageNet classification with deep convolutional neural networks. Proceedings of the 25th International Conference on Neural Information Processing Systems. Lake Tahoe, NV, 1097–1105Google Scholar
- Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. 2016. Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, Las Vegas, NV, 770–778. DOI: 10.1109/CVPR.2016.90Google ScholarCross Ref
- Zhe Xu, Xiaomin Zhao, Xi Guo, and Jiaxin Guo. 2019. Deep learning application for predicting soil organic matter content by VIS-NIR spectroscopy. Computational Intelligence and Neuroscience 2019 (November 2019), Article ID 3563761, 11 pages. https://doi.org/10.1155/2019/3563761Google Scholar
- Lei Hou, QingXiang Wu, Qiyan Sun, Heng Yang, and Pengfei Li. 2016. Fruit recognition based on convolution neural network. 2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD). Changsha, 18–22, DOI: 10.1109/FSKD.2016.7603144Google ScholarCross Ref
- Zaidah Ibrahim, Nurbaity Sabri, and Dino Isa. 2018. Palm oil fresh fruit bunch ripeness grading recognition using convolutional neural network. Journal of Telecommunication, Electronic and Computer Engineering (JTEC) 10, 3-2 (September 2018), 109–113Google Scholar
- DLP NIRscan Nano EVM User's Guide. Retrieved June 21, 2020 from https://www.ti.com/lit/ug/dlpu030g/dlpu030g.pdf?ts=1592754432447&ref_url=https%253A%252F%252Fwww.google.com%252FGoogle Scholar
- Duy K. Ninh, Masanori Morise, and Yoichi Yamashita. 2013. A generation error function considering dynamic properties of speech parameters for minimum generation error training for hidden Markov model-based speech synthesis. Acoustical Science and Technology 34, 2 (February 2013), 123–132. https://doi.org/10.1250/ast.34.123Google ScholarCross Ref
- Abraham Savitzky and Marcel J. E. Golay. 1964. Smoothing and differentiation of data by simplified least squares procedures. Analytical Chemistry 36, 8 (July 1964), 1627–1639. https://doi.org/10.1021/ac60214a047Google ScholarCross Ref
Recommendations
Peach variety identification using near-infrared diffuse reflectance spectroscopy
NIR spectroscopy was successfully used to identify peach varieties.Peach variety identification models established on PCA reached 100% accuracy.LSSVM performed better than ELM in identifying peach varieties.NIR spectroscopy had potential in developing ...
Fruits and Vegetables Calorie Counter Using Convolutional Neural Networks
DH '16: Proceedings of the 6th International Conference on Digital Health ConferencePeople care about what types of fruit they're eating and the nutrients it contains because eating fruit is an essential part of leading a healthy life. This paper introduces an automatic way for detecting and recognizing the fruits in an image. The ...
Vision-based fruit recognition via multi-scale attention CNN
AbstractFruit quality assessment, grading and sorting are of vital importance to fruit processing, and all these involve fruit recognition. Vision-based fruit recognition can recognize fruit automatically and further support more applications ...
Highlights- We adopt a Multi-Scale Attention Network (MSANet) for fruit recognition.
- ...
Comments