Defect segmentation of apples is an important task in the agriculture industry for quality control and food safety. In this paper, we propose a deep learning approach for the automated segmentation of apple defects using convolutional neural networks (CNNs) based on a U-shaped architecture with skip-connections only within the noise reduction block. An ad-hoc data synthesis technique has been designed to increase the number of samples and at the same time to reduce neural network overfitting. We evaluate our model on a dataset of multi-spectral apple images with pixel-wise annotations for several types of defects. In this paper, we show that our proposal outperforms in terms of segmentation accuracy general-purpose deep learning architectures commonly used for segmentation tasks. From the application point of view, we improve the previous methods for apple defect segmentation. A measure of the computational cost shows that our proposal can be employed in real-time (about 100 frame-per-second on GPU) and in quasi-real-time (about 7/8 frame-per-second on CPU) visual-based apple inspection. To further improve the applicability of the method, we investigate the potential of using only RGB images instead of multi-spectral images as input images. The results prove that the accuracy in this case is almost comparable with the multi-spectral case.

Agarla, M., Napoletano, P., Schettini, R. (2023). Quasi Real-Time Apple Defect Segmentation Using Deep Learning. SENSORS, 23(18) [10.3390/s23187893].

Quasi Real-Time Apple Defect Segmentation Using Deep Learning

Agarla, M;Napoletano, P;Schettini, R
2023

Abstract

Defect segmentation of apples is an important task in the agriculture industry for quality control and food safety. In this paper, we propose a deep learning approach for the automated segmentation of apple defects using convolutional neural networks (CNNs) based on a U-shaped architecture with skip-connections only within the noise reduction block. An ad-hoc data synthesis technique has been designed to increase the number of samples and at the same time to reduce neural network overfitting. We evaluate our model on a dataset of multi-spectral apple images with pixel-wise annotations for several types of defects. In this paper, we show that our proposal outperforms in terms of segmentation accuracy general-purpose deep learning architectures commonly used for segmentation tasks. From the application point of view, we improve the previous methods for apple defect segmentation. A measure of the computational cost shows that our proposal can be employed in real-time (about 100 frame-per-second on GPU) and in quasi-real-time (about 7/8 frame-per-second on CPU) visual-based apple inspection. To further improve the applicability of the method, we investigate the potential of using only RGB images instead of multi-spectral images as input images. The results prove that the accuracy in this case is almost comparable with the multi-spectral case.
Articolo in rivista - Articolo scientifico
apple defect segmentation; multispectral imaging; real-time deep learning; visual inspection;
English
14-set-2023
2023
23
18
7893
open
Agarla, M., Napoletano, P., Schettini, R. (2023). Quasi Real-Time Apple Defect Segmentation Using Deep Learning. SENSORS, 23(18) [10.3390/s23187893].
File in questo prodotto:
File Dimensione Formato  
10281-467295_VoR.pdf

accesso aperto

Tipologia di allegato: Publisher’s Version (Version of Record, VoR)
Licenza: Creative Commons
Dimensione 6.47 MB
Formato Adobe PDF
6.47 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/467295
Citazioni
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 1
Social impact