Promotion of Color Sorting in Industrial Systems Using a Deep Learning Algorithm
Apstrakt
Color sorting is a technological operation performed with the aim of classifying compliant and noncompliant agricultural products in large-capacity industrial systems for agricultural product processing. This paper investigates the application of the YOLOv3 algorithm on raspberry images as a method developed for the detection, localization, and classification of objects based on convolutional neural networks (CNNs). To our knowledge, this is the first time a YOLO algorithm or CNN has been used with original images from the color sorter to focus on agricultural products. Results of the F1 measure were in the 92–97% range. Images in full resolution, 1024 × 1024, produced an average detection time of 0.37 s. The impact of the hyperparameters that define the YOLOv3 model as well as the impact of the application of the chosen augmentative methods on the model are evaluated. The successful classification of stalks, which is particularly challenging due to their shape, small dimensions, and v...ariations, was achieved. The presented model demonstrates the ability to classify noncompliant products into four classes, some of which are appropriate for reprocessing. The software, including a graphic interface that enables the real-time testing of machine learning algorithm, is developed and presented.
Ključne reči:
machine learning / deep learning / YOLO / color sorting / inspection / agricultureIzvor:
Aplied Sciences, 2022, 12, 24, 12817-Izdavač:
- MDPI
Finansiranje / projekti:
- Ministarstvo nauke, tehnološkog razvoja i inovacija Republike Srbije, institucionalno finansiranje - 200105 (Univerzitet u Beogradu, Mašinski fakultet) (RS-MESTD-inst-2020-200105)
Institucija/grupa
Mašinski fakultetTY - JOUR AU - Medojević, Ivana AU - Veg, Emil AU - Joksimović, Aleksandra AU - Ilić, Jelena PY - 2022 UR - https://machinery.mas.bg.ac.rs/handle/123456789/4477 AB - Color sorting is a technological operation performed with the aim of classifying compliant and noncompliant agricultural products in large-capacity industrial systems for agricultural product processing. This paper investigates the application of the YOLOv3 algorithm on raspberry images as a method developed for the detection, localization, and classification of objects based on convolutional neural networks (CNNs). To our knowledge, this is the first time a YOLO algorithm or CNN has been used with original images from the color sorter to focus on agricultural products. Results of the F1 measure were in the 92–97% range. Images in full resolution, 1024 × 1024, produced an average detection time of 0.37 s. The impact of the hyperparameters that define the YOLOv3 model as well as the impact of the application of the chosen augmentative methods on the model are evaluated. The successful classification of stalks, which is particularly challenging due to their shape, small dimensions, and variations, was achieved. The presented model demonstrates the ability to classify noncompliant products into four classes, some of which are appropriate for reprocessing. The software, including a graphic interface that enables the real-time testing of machine learning algorithm, is developed and presented. PB - MDPI T2 - Aplied Sciences T1 - Promotion of Color Sorting in Industrial Systems Using a Deep Learning Algorithm IS - 24 SP - 12817 VL - 12 DO - 10.3390/app122412817 ER -
@article{ author = "Medojević, Ivana and Veg, Emil and Joksimović, Aleksandra and Ilić, Jelena", year = "2022", abstract = "Color sorting is a technological operation performed with the aim of classifying compliant and noncompliant agricultural products in large-capacity industrial systems for agricultural product processing. This paper investigates the application of the YOLOv3 algorithm on raspberry images as a method developed for the detection, localization, and classification of objects based on convolutional neural networks (CNNs). To our knowledge, this is the first time a YOLO algorithm or CNN has been used with original images from the color sorter to focus on agricultural products. Results of the F1 measure were in the 92–97% range. Images in full resolution, 1024 × 1024, produced an average detection time of 0.37 s. The impact of the hyperparameters that define the YOLOv3 model as well as the impact of the application of the chosen augmentative methods on the model are evaluated. The successful classification of stalks, which is particularly challenging due to their shape, small dimensions, and variations, was achieved. The presented model demonstrates the ability to classify noncompliant products into four classes, some of which are appropriate for reprocessing. The software, including a graphic interface that enables the real-time testing of machine learning algorithm, is developed and presented.", publisher = "MDPI", journal = "Aplied Sciences", title = "Promotion of Color Sorting in Industrial Systems Using a Deep Learning Algorithm", number = "24", pages = "12817", volume = "12", doi = "10.3390/app122412817" }
Medojević, I., Veg, E., Joksimović, A.,& Ilić, J.. (2022). Promotion of Color Sorting in Industrial Systems Using a Deep Learning Algorithm. in Aplied Sciences MDPI., 12(24), 12817. https://doi.org/10.3390/app122412817
Medojević I, Veg E, Joksimović A, Ilić J. Promotion of Color Sorting in Industrial Systems Using a Deep Learning Algorithm. in Aplied Sciences. 2022;12(24):12817. doi:10.3390/app122412817 .
Medojević, Ivana, Veg, Emil, Joksimović, Aleksandra, Ilić, Jelena, "Promotion of Color Sorting in Industrial Systems Using a Deep Learning Algorithm" in Aplied Sciences, 12, no. 24 (2022):12817, https://doi.org/10.3390/app122412817 . .