Abstract
This article presents a novel approach to identify two species of fruit insect pests as part of a network of intelligent traps designed to monitor the population of these insects in a plantation. The proposed approach uses a simple Digital Image Processing technique to detect regions in the image that are likely the monitored pests and an Artificial Neural Network to classify the regions into the right class given their characteristics. This identification is done essentially by a Convolutional Neural Network (CNN), which learns the characteristics of the insects based on their images made from the adhesive floor inside a trap. We have trained several CNN architectures, with different configurations, through a data set of images collected in the field. We aimed to find the model with the highest precision and the lowest time needed for the classification. The best performance in classification was achieved by ResNet18, with a precision of 93.55% and 91.28% for the classification of the pests focused on this study, named Ceratitis capitata and Grapholita molesta, respectively, and a 90.72%overall accuracy. Yet, the classification must be embedded on a resource-constrained system inside the trap, then we exploited SqueezeNet, MobileNet, and MNASNet architectures to achieve a model with lesser inference time and small losses in accuracy when compared to the models we assessed. We also attempted to quantize our highest precision model to reduce even more inference time in embedded systems, which achieved a precision of 88.76% and 89.73% for C. capitata and G. molesta, respectively; notwithstanding, a decrease of roughly 2% on the overall accuracy was endured. According to the expertise of our partner company, our results are worthwhile for a real-world application, since general human laborers have a precision of about 85%.
- [1] E. Goldshtein, Y. Cohen, A. Hetzroni, Y. Gazit, D. Timar, L. Rosenfeld, Y. Grinshpon, A. Hoffman, and A. Mizrach. 2017. Development of an automatic monitoring trap for Mediterranean fruit fly (Ceratitis capitata) to optimize control applications frequency. Comput. Electron. Agric. 139 (2017), 115–125. Google ScholarCross Ref
- [2] . 2008. A unified architecture for natural language processing: Deep neural networks with multitask learning. In Proceedings of the 25th International Conference on Machine Learning. ACM, 160–167.Google ScholarDigital Library
- [3] . 2013. Nível de infestação de moscas-das-frutas em faixa de fronteira, no Rio Grande do Sul. Revista Ceres 60 (Aug. 2013), 589–593. Retrieved from http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0034-737X2013000400020&nrm=iso.Google ScholarCross Ref
- [4] . 2016. Automatic moth detection from trap images for pest management. Comput. Electron. Agric. 123 (
04 2016), 17–28. Google ScholarDigital Library - [5] . 2020. QNNPACK: Open Source Library for Optimized Mobile Deep Learning. Retrieved from https://engineering.fb.com/ml-applications/qnnpack/.Google Scholar
- [6] . 1995. Integrated Pest Management. Springer Science & Business Media, Berlim, Alemanha.Google Scholar
- [7] . 2014. Automatic identification of fruit flies (Diptera: Tephritidae). J. Visual Commun. Image Represent. 25, 7 (2014), 1516–1527.Google ScholarDigital Library
- [8] . 2011. Deep sparse rectifier neural networks. In Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics. 315–323.Google Scholar
- [9] . 2014. Compressing deep convolutional networks using vector quantization. Retrieved from http://arxiv.org/abs/1412.6115.Google Scholar
- [10] . 2016. Deep Learning. MIT Press. Retrieved from http://www.deeplearningbook.org.Google ScholarDigital Library
- [11] . 2016. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR’16). 770–778. Google ScholarCross Ref
- [12] . 2018. AMC: Automated deep compression and acceleration with reinforcement learning. Retrieved from http://arxiv.org/abs/1802.03494.Google Scholar
- [13] . 2016. Densely connected convolutional networks. Retrieved from http://arxiv.org/abs/1608.06993.Google Scholar
- [14] . 2017. SqueezeNet: AlexNet-level accuracy with \( 50\times \) fewer parameters and <1MB model size. Retrieved from https://arxiv.org/abs/1602.07360.Google Scholar
- [15] . 2015. Batch normalization: Accelerating deep network training by reducing internal covariate shift. Retrieved from http://arxiv.org/abs/1502.03167.Google Scholar
- [16] . 2018. Quantization and training of neural networks for efficient integer-arithmetic-only inference. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2704–2713.Google ScholarCross Ref
- [17] . 2008. A GSM-based remote wireless automatic monitoring system for field information: A case study for ecological monitoring of the oriental fruit fly, Bactrocera dorsalis (Hendel). Comput. Electron. Agric. 62, 2 (2008), 243–259. Google ScholarDigital Library
- [18] . 2018. Quantizing deep convolutional networks for efficient inference: A whitepaper. Retrieved from http://arxiv.org/abs/1806.08342.Google Scholar
- [19] . 2012. Imagenet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems. MIT Press, 1097–1105.Google ScholarDigital Library
- [20] . 2010. Robust insect classification applied to real time greenhouse infestation monitoring. In Proceedings of the 20th International Conference on Pattern Recognition on Visual Observation and Analysis of Animal and Insect Behavior Workshop. 1–4.Google Scholar
- [21] . 1989. Backpropagation applied to handwritten zip code recognition. Neural Comput. 1, 4 (1989), 541–551.Google ScholarDigital Library
- [22] . 2013. Network in network. Retrieved from https://arxiv.org/abs/1312.4400.Google Scholar
- [23] . 2015. Fully convolutional networks for semantic segmentation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 3431–3440.Google ScholarCross Ref
- [24] . 2012. Monitoring pest insect traps by means of low-power image sensor technologies. Sensors 12, 11 (2012), 15801–15819.Google ScholarCross Ref
- [25] . 2014. Learning and transferring mid-level image representations using convolutional neural networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 1717–1724.Google ScholarDigital Library
- [26] . 2015. Deep face recognition. In Proceedings of the British Machine Vision Conference (BMVC’15), Vol. 1, 6.Google ScholarCross Ref
- [27] . 2019. PyTorch: An imperative style, high-performance deep learning library. In Advances in Neural Information Processing Systems 32, , , , , , and (Eds.). Curran Associates, 8024–8035. Retrieved from http://papers.neurips.cc/paper/9015-pytorch-an-imperative-style-high-performance-deep-learning-library.pdf.Google Scholar
- [28] . 2017. The effectiveness of data augmentation in image classification using deep learning. Retrieved from http://arxiv.org/abs/1712.04621.Google Scholar
- [29] . 2017. Automated remote insect surveillance at a global scale and the Internet of Things. Robotics 6, 3 (2017). Google ScholarCross Ref
- [30] . 2011. Evaluation: From precision, recall and F-measure to ROC, informedness, markedness and correlation. J. Mach. Learn. Technol. 2 (Jan. 2011), 2229–3981. Google ScholarCross Ref
- [31] 2018. Identification of fruit fly in intelligent traps using techniques of digital image processing and machine learning. In Proceedings of the 33rd Annual ACM Symposium on Applied Computing (SAC’18). ACM, New York, NY, 260–267. Google ScholarDigital Library
- [32] . 1958. The perceptron: A probabilistic model for information storage and organization in the brain.Psychol. Rev. 65, 6 (1958), 386.Google ScholarCross Ref
- [33] . 1961. Principles of Neurodynamics: Perceptrons and the Theory of Brain Mechanisms.
Technical Report , Cornell Aeronautical Lab, Buffalo, NY.Google ScholarCross Ref - [34] . 2018. Inverted residuals and linear bottlenecks: Mobile networks for classification, detection and segmentation. Retrieved from http://arxiv.org/abs/1801.04381.Google Scholar
- [35] . 2006. Concepts and applications of trap cropping in pest management. Annu. Rev. Entomol. 51, 1 (2006), 285–308.Google ScholarCross Ref
- [36] . 2009. Machine vision algorithm for whiteflies (Bemisia tabaci Genn.) scouting under greenhouse environment. J. Appl. Entomol. 133, 7 (2009), 546–552.Google ScholarCross Ref
- [37] . 2019. And the bit goes down: Revisiting the quantization of neural networks. Retrieved from http://arxiv.org/abs/1907.05686.Google Scholar
- [38] . 2018. MnasNet: Platform-aware neural architecture search for Mobile. Retrieved from http://arxiv.org/abs/1807.11626.Google Scholar
- [39] . 2011. Automatic monitoring of pest insects traps by Zigbee-based wireless networking of image sensors. In Proceedings of the Instrumentation and Measurement Technology Conference (I2MTC’11). IEEE, 1–5.Google ScholarCross Ref
- [40] . 2012. Image-based orchard insect automated identification and classification method. Comput. Electron. Agric. 89 (Nov. 2012), 110–115. Google ScholarDigital Library
- [41] . 2017. Aggregated residual transformations for deep neural networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR’17). 5987–5995. Google ScholarCross Ref
- [42] . 2018. ShuffleNet: An extremely efficient convolutional neural network for mobile devices. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 6848–6856.Google ScholarCross Ref
Index Terms
- Deep Learning Embedded into Smart Traps for Fruit Insect Pests Detection
Recommendations
Intelligent detection of citrus fruit pests using machine vision system and convolutional neural network through transfer learning technique
AbstractPlant pests and diseases play a significant role in reducing the quality of agricultural products. As one of the most important plant pathogens, pests like Mediterranean fruit fly cause significant damage to crops and thus annually farmers face a ...
Highlights- The detection accuracy of infected fruits with Mediterranean larvae was obtained using transfer learning technique 99.33%.
- Performance comparisons between four CNN pre-trained models for detecting infected fruits were examined.
- The ...
Image Recognition of Crop Diseases and Insect Pests Based on Deep Learning
Deep learning algorithms have the advantages of clear structure and high accuracy in image recognition. Accurate identification of pests and diseases in crops can improve the pertinence of pest control in farmland, which is beneficial to agricultural ...
Reflectance-based assessment of spider mite "bio-response" to maize leaves and plant potassium content in different irrigation regimes
It is widely accepted that pest infestations elicit a change in plant physiology, which cause detectable changes in crop leaf reflectance. In this study, we test the hypothesis that crop leaf reflectance may also be used to forecast the risk of pest ...
Comments