journal article·2020-04-22Open access

EmbeddedPigDet—Fast and Accurate Pig Detection for Embedded Board Implementations

Jihyun Seo, Hanse Ahn, Daewon Kim, Sungju Lee, Yongwha Chung, Daihee Park

Applied Sciences· Volume 10· Issue 8

Abstract

Automated pig monitoring is an important issue in the surveillance environment of a pig farm. For a large-scale pig farm in particular, practical issues such as monitoring cost should be considered but such consideration based on low-cost embedded boards has not yet been reported. Since low-cost embedded boards have more limited computing power than typical PCs and have tradeoffs between execution speed and accuracy, achieving fast and accurate detection of individual pigs for “on-device” pig monitoring applications is very challenging. Therefore, in this paper, we propose a method for the fast detection of individual pigs by reducing the computational workload of 3 × 3 convolution in widely-used, deep learning-based object detectors. Then, in order to recover the accuracy of the “light-weight” deep learning-based object detector, we generate a three-channel composite image as its input image, through “simple” image preprocessing techniques. Our experimental results on an NVIDIA Jetson Nano embedded board show that the proposed method can improve the integrated performance of both execution speed and accuracy of widely-used, deep learning-based object detectors, by a factor of up to 8.7.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Industrial Vision Systems and Defect Detection · Video Surveillance and Tracking Methods · Advanced Neural Network Applications

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

Department of Computer Convergence Software, Korea University, Sejong 30019, Korea

Department of Software, Sangmyung University, Cheonan 31066, Korea

References (94)

Show source-provided references
1. Precision Livestock Farming: An International Review of Scientific and Commercial Aspects
2. Recent Advances in Wearable Sensors for Animal Health Management10.1016/j.sbsr.2016.11.004 ↗
3. Tullo, E., Fontana, I., and Guarino, M. (2013, January 10–12). Precision livestock farming: An overview of image and sound labelling. Proceedings of the 6th European Conference on Precision Livestock Farming, Leuven, Belgium.
4. Early Detection of Health and Welfare Compromises through Automated Detection of Behavioural Changes in Pigs10.1016/j.tvjl.2016.09.005 ↗
5. A Brief Review of the Application of Machine Vision in Livestock Behaviour Analysis
6. Korean Government (2016). 4th Industrial Revolution and Agriculture, (In Korean).
7. Han, S., Zhang, J., Zhu, M., Wu, J., and Kong, F. (2017, January 26–28). Review of automatic detection of pig behaviours by using Image Analysis. Proceedings of the International Conference on AEECE, Chengdu, China.10.1088/1755-1315/69/1/012096 ↗
8. Evaluation of Image Analysis as A Means of Estimating the Weight of Pigs10.1016/0021-8634(90)80048-Y ↗
9. Image-Analysis Parameters as Inputs for Automatic Environmental Temperature Control in Piglet Houses10.1016/0168-1699(90)90014-G ↗
10. Segmentation and Tracking of Piglets in Images10.1007/BF01215814 ↗
11. The Automated Analysis of Clustering Behaviour of Piglets from Thermal Images in response to Immune Challenge by Vaccination10.1017/S1751731117001239 ↗
12. Model-based Detection of Pigs in Images under Sub-Optimal Conditions10.1016/j.compag.2018.06.043 ↗
13. Illumination and Reflectance Estimation with its Application in Foreground10.3390/s150921407 ↗
14. Segmentation of Sows in Farrowing Pens10.1049/iet-ipr.2012.0734 ↗
15. Foreground Detection using Loopy Belief Propagation10.1016/j.biosystemseng.2013.06.011 ↗
16. Development of Automatic Surveillance of Animal Behaviour and Welfare using Image Analysis and Machine Learned Segmentation Techniques10.1017/S1751731115001342 ↗
17. Automatic Monitoring of Pig Locomotion using Image Analysis10.1016/j.livsci.2013.11.007 ↗
18. Automatic Estimation of Number of Piglets in a Pen during Farrowing, using Image Analysis10.1016/j.biosystemseng.2016.08.018 ↗
19. Development of a Real-Time Computer Vision System for Tracking Loose-Housed Pigs10.1016/j.compag.2011.01.011 ↗
20. Real-Time Recognition of Sows in Video: A Supervised Approach
21. Automatic Detection of Mounting Behaviours among Pigs using Image Analysis10.1016/j.compag.2016.04.022 ↗
22. A New Approach for Categorizing Pig Lying Behaviour based on a Delaunay Triangulation Method10.1017/S1751731116001208 ↗
23. Using Automated Image Analysis in Pig Behavioural Research: Assessment of the Influence of Enrichment Subtrate Provision on Lying Behaviour10.1016/j.applanim.2017.06.015 ↗
24. Foreground Detection of Group-Housed Pigs based on the Combination of Mixture of Gaussians using Prediction Mechanism and Threshold Segmentation10.1016/j.biosystemseng.2014.07.002 ↗
25. Multi-Object Extraction from Topview Group-Housed Pig Images based on Adaptive Partitioning and Multilevel Thresholding Segmentation10.1016/j.biosystemseng.2015.05.001 ↗
26. Boundary Detection of Pigs in Pens based on Adaptive Thresholding using an Integral Image and Adaptive Partitioning
27. An Automatic Splitting Method for the Adhesive Piglets Gray Scale Image based on the Ellipse Shape Feature10.1016/j.compag.2015.11.008 ↗
28. An Automatic Ear Base Temperature Extraction Method for Top View Piglet Thermal Image10.1016/j.compag.2018.10.030 ↗
29. Estimating Pig Weights from Images without Constraint on Posture and Illumination10.1016/j.compag.2018.08.006 ↗
30. Kang, F., Wang, C., Li, J., and Zong, Z. (2018). A Multiobjective Piglet Image Segmentation Method based on an Improved Noninteractive GrabCut Algorithm. Adv. Multimed., 108876.10.1155/2018/1083876 ↗
31. Automatic Recognition of Sow Nursing Behavious using Deep Learning-based Segmentation and Spatial and Temporal Features10.1016/j.biosystemseng.2018.09.011 ↗
32. Feeding Behavior Recognition for Group-Housed Pigs with the Faster R-CNN10.1016/j.compag.2018.11.002 ↗
33. Estimation of Pig Weight using a Microsoft Kinect Prototype Imaging System10.1016/j.compag.2014.08.008 ↗
34. Automatic Recognition of Lactating Sow Behaviors through Depth Image Processing10.1016/j.compag.2016.04.026 ↗
35. Validity of the Microsoft Kinect Sensor for Assessment of Normal Walking Patterns in Pigs10.1016/j.compag.2015.07.003 ↗
36. Zhu, Q., Ren, J., Barclay, D., McCormack, S., and Thomson, W. (2015, January 21–23). Automatic animal detection from kinect sensed images for livestock monitoring and assessment. Proceedings of the International Conference on Computer and Information Technology, ICCCIT, Dhaka, Bangladesh.10.1109/CIT/IUCC/DASC/PICOM.2015.172 ↗
37. Application of 3D Imaging Sensor for Tracking Minipigs in the Open Field Test10.1016/j.jneumeth.2014.07.012 ↗
38. An Approach of Pig Weight Estimation using Binocular Stereo System based on LabVIEW10.1016/j.compag.2016.08.012 ↗
39. Automated Tracking to Measure Behavioural Changes in Pigs for Health and Welfare Monitoring10.1038/s41598-017-17451-6 ↗
40. Automatic Recognition of Lactating Sow Postures from Depth Images by Deep Learning Detector10.1016/j.compag.2018.01.023 ↗
41. Lee, J., Jin, L., Park, D., and Chung, Y. (2016). Automatic Recognition of Aggressive Pig Behaviors using Kinect Depth Sensor. Sensors, 16.10.3390/s16050631 ↗
42. Kim, J., Chung, Y., Choi, Y., Sa, J., Kim, H., Chung, Y., Park, D., and Kim, H. (2017). Depth-based Detection of Standing-Pigs in Moving Noise Environments. Sensors, 17.10.3390/s17122757 ↗
43. A Cost-Effective Pigsty Monitoring System based on a Video Sensor
44. Sa, J., Choi, Y., Lee, H., Chung, Y., Park, D., and Cho, J. (2019). Fast Pig Detection with a Topview Camera under Various Illumination Conditions. Symmetry, 11.10.3390/sym11020266 ↗
45. Zhang, L., Gray, H., Ye, X., Collins, L., and Allinson, N. (2019). Automatic Individual Pig Detection and Tracking in Pig Farms. Sensors, 19.10.3390/s19051188 ↗
46. Automatic Scoring of Lateral and Sternal Lying Posture in Grouped Pigs Using Image Processing and Support Vector Machine10.1016/j.compag.2018.12.009 ↗
47. Psota, E., Mittek, M., Perez, L., Schmidt, T., and Mote, B. (2019). Multi-Pig Part Detection and Association with a Fully-Convolutional Network. Sensors, 19.10.3390/s19040852 ↗
48. Group-Housed Pig Detection in Video Surveillance of Overhead Views using Multi-Feature Template Matching10.1016/j.biosystemseng.2019.02.018 ↗
49. Practical Monitoring of Undergrown Pigs for IoT-Based Large-Scale Smart Farm10.1109/ACCESS.2019.2955761 ↗
50. NVIDIA (2019, November 10). NVIDIA Jetson Nano. Available online: http://www.nvidia.com/object/embedded-systems-dev-kits-modules.html.
51. Machine Learning for Internet of Things Data Analysis: A Survey10.1016/j.dcan.2017.10.002 ↗
52. Ham, M., Moon, J., Lim, G., Song, W., Jung, J., Ahn, H., Woo, S., Cho, Y., Park, J., and Oh, S. (2019). NNStreamer: Stream Processing Paradigm for Neural Networks, Toward Efficient Development and Execution of On-Device AI Applications. arXiv.
53. Nguyen, P., Arsalan, M., Koo, J., Naqvi, R., Truong, N., and Park, K. (2018). LightDenseYOLO: A Fast and Accurate Marker Tracker for Autonomous UAV Landing by Visible Light Camera Sensor on Drone. Sensors, 18.10.3390/s18061703 ↗
54. Xiao, J., Wu, H., and Li, X. (2019). Internet of Things Meets Vehicles: Sheltering In-Vehicle Network through Lightweight Machine Learning. Symmetry, 11.10.3390/sym11111388 ↗
55. Yang, T., Howard, A., Chen, B., Zhang, X., Go, A., Sandler, M., Sze, V., and Adam, H. (2018, January 8–14). Netadapt: Platform-aware neural network adaptation for mobile applications. Proceedings of the European Conference on Computer Vision, Munich, Germany.10.1007/978-3-030-01249-6_18 ↗
56. Intel (2018, February 28). Intel RealSense D435. Available online: https://click.intel.com/intelr-realsensetm-depth-camera-d435.html.
57. Adaptive Thresholding using the Integral Image10.1080/2151237X.2007.10129236 ↗
58. Redmon, J., Divvala, S., Girshick, R., and Farhadi, A. (July, January 26). You only look once: Unified, real-time object detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vega, NV, USA.
59. Girshick, R., Donahue, J., Darrell, T., and Malik, J. (2014, January 24–27). Rich feature hierarchies for accurate object detection and semantic segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA.10.1109/CVPR.2014.81 ↗
60. Girshick, R. (2015, January 11–18). Fast R-CNN. Proceedings of the IEEE International Conference on Computer Vision, Las Condes, Chile.10.1109/ICCV.2015.169 ↗
61. Ren, S., He, K., Girshick, R., and Sun, J. (2015, January 7–12). Faster R-CNN: Towards real-time object detection with region proposal networks. Proceedings of the Advances in Neural Information Processing Systems, Montréal, QC, Canada.
62. Automated video analysis of pig activity at pen level highly correlates to human observations of behavioural activities10.1016/j.livsci.2013.12.011 ↗
63. Recognition of aggressive episodes of pigs based on convolutional neural network and long short-term memory10.1016/j.compag.2019.105166 ↗
64. Detection of aggressive behaviours in pigs using a RealSence depth sensor10.1016/j.compag.2019.105003 ↗
65. Li, H., Kadav, A., Durdanovic, I., Samet, H., and Graf, H.P. (2016). Pruning filters for efficient convnets. arXiv.
66. He, Y., Kang, G., Dong, X., Fu, Y., and Yang, Y. (2018). Soft filter pruning for accelerating deep convolutional neural networks. arXiv.10.24963/ijcai.2018/309 ↗
67. Yu, R., Li, A., Chen, C., Lai, J., Morariu, V., Han, X., and Davis, L. (2018, January 18–23). Nisp: Pruning networks using neuron importance score propagation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.10.1109/CVPR.2018.00958 ↗
68. Ding, X., Ding, G., Han, J., and Tang, S. (2018, January 2–7). Auto-balanced filter pruning for efficient convolutional neural networks. Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence, New Orleans, LA, USA.10.1609/aaai.v32i1.12262 ↗
69. Lin, S., Ji, R., Li, Y., Wu, Y., Huang, F., and Zhang, B. (2018, January 13–19). Accelerating convolutional networks via global & dynamic filter pruning. Proceedings of the International Joint Conferences on Artificial Intelligence, Stockholm, Sweden.10.24963/ijcai.2018/336 ↗
70. Peng, B., Tan, W., Li, Z., Zhang, S., Xie, D., and Pu, S. (2018, January 8–14). Extreme network compression via filter group approximation. Proceedings of the European Conference on Computer Vision, Munich, Germany.10.1007/978-3-030-01237-3_19 ↗
71. Zhuang, Z., Tan, M., Zhuang, B., Liu, J., Guo, Y., Wu, Q., and Zhu, J. (2018, January 3–8). Discrimination-aware channel pruning for deep neural networks. Proceedings of the Advances in Neural Information Processing Systems, Montreal, QC, Canada.
72. He, K., Zhang, X., Ren, S., and Sun, J. (2016, January 27–30). Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA.10.1109/CVPR.2016.90 ↗
73. Zuiderveld, K. (1994). Contrast Limited Adaptive Histogram Equalization, Academic Press Inc.10.1016/B978-0-12-336156-1.50061-6 ↗
74. Redmon, J., and Farhadi, A. (2017, January 21–26). YOLO9000: Better, faster, stronger. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.10.1109/CVPR.2017.690 ↗
75. Redmon, J., and Farhadi, A. (2018). Yolov3: An incremental improvement. arXiv.
76. OpenCV (2019, April 30). Open Source Computer Vision. Available online: http://opencv.org.
77. NVIDIA (2019, April 30). NVIDIA Jetson TX2. Available online: http://www.nvidia.com/object/embedded-systems-dev-kits-modules.html.
78. Lin, T., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollar, P., and Zitnick, C. (2014, January 6–12). Microsoft COCO: Common Objects in Context. Proceedings of the European Conference on Computer Vision, Zurich, Switzerland.10.1007/978-3-319-10602-1_48 ↗
79. The Pascal Visual Object Classes (VOC) Challenge10.1007/s11263-009-0275-4 ↗
80. The Impact of the PCA Dimensionality Reduction for CNN based Hyperspectral Image Classification
81. Pig Detection Algorithm Based on Sliding Windows and PCA Convolution10.1109/ACCESS.2019.2907748 ↗
82. Kim, J., Choi, Y., Sa, J., Ju, M., Chung, Y., Park, D., and Kim, H. (2016). Pig Detection Using Texture Information, The Institute of Electronics and Information Engineers. (In Korean).10.1145/3127942.3127949 ↗
83. Noise-Robust Porcine Respiratory Diseases Classification Using Texture Analysis and CNN
84. Learning motion features for example-based finger motion estimation for virtual characters10.1007/s13319-017-0136-9 ↗
85. Yuan, P., Zhong, Y., and Yuan, Y. (2017). Faster r-cnn with region proposal refinement. Tech. Rep.
86. Han, W., Khorrami, P., Paine, T., Ramachandran, P., Babaeizadeh, M., Shi, H., and Huang, T. (2016). Seq-nms for video object detection. arXiv.
87. Zhou, Y., Li, Z., Xiao, S., He, C., Huang, Z., and Li, H. (2018). Auto-conditioned recurrent networks for extended complex human motion synthesis. arXiv.
88. Ngiam, J., Khosla, A., Kim, M., Nam, J., Lee, H., and Ng, A. (July, January 28). Multimodal deep learning. Proceedings of the International Conference on Machine Learning, Washington, DC, USA.
89. Automatic detection and recognition of pig wasting diseases using sound data in audio surveillance systems10.3390/s131012929 ↗
90. Pigs boundary detection using both color and depth information
91. Kim, H. (2015, January 18–19). Automatic identification of a coughing animal using audio and video data. Proceedings of the Fourth International Conference on Information Science and Cloud Computing, Guangzhou, China.
92. Bai, J., Zhang, H., and Li, Z. (2018, January 28–30). The generalized detection method for the dim small targets by faster R-CNN integrated with GAN. Proceedings of the IEEE 3rd International Conference on Communication and Information Systems (ICCIS), Singapore.10.1109/ICOMIS.2018.8644960 ↗
93. Enhanced Sound Signal Based Sound-Event Classification
94. Novel unsupervised auditory filterbank learning using convolutional RBM for speech recognition10.1109/TASLP.2016.2607341 ↗