TY - JOUR
T1 - Automated Apple Recognition System Using Semantic Segmentation Networks with Group and Shuffle Operators
AU - Zulkifley, Mohd Asyraf
AU - Moubark, Asraf Mohamed
AU - Saputro, Adhi Harmoko
AU - Abdani, Siti Raihanah
N1 - Funding Information:
Funding: This research was funded by Universiti Kebangsaan Malaysia under Dana padanan Kolaborasi with a grant number DPK-2021-012 and Ministry of Higher Education Malaysia with grant number FRGS/1/2019/ICT02/UKM/02/1.
Publisher Copyright:
© 2022 by the authors. Licensee MDPI, Basel, Switzerland.
PY - 2022/6
Y1 - 2022/6
N2 - Apples are one of the most consumed fruits, and they require efficient harvesting proce-dures to remains in optimal states for a longer period, especially during transportation. Therefore, automation has been adopted by many orchard operators to help in the harvesting process, which includes apple localization on the trees. The de facto sensor that is currently used for this task is the standard camera, which can capture wide view information of various apple trees from a reasonable distance. Therefore, this paper aims to produce the output mask of the apple locations on the tree automatically by using a deep semantic segmentation network. The network must be robust enough to overcome all challenges of shadow, surrounding illumination, size variations, and occlusion to produce accurate pixel-wise localization of the apples. A high-resolution deep architecture is em-bedded with an optimized design of group and shuffle operators (GSO) to produce the best apple segmentation network. GSO allows the network to reduce the dependency on a few sets of dominant convolutional filters by forcing each smaller group to contribute effectively to the task of extracting optimal apple features. The experimental results show that the proposed network, GSHR-Net, with two sets of group convolution applied to all layers produced the best mean intersection over union of 0.8045. The performance has been benchmarked with 11 other state-of-the-art deep semantic segmentation networks. For future work, the network performance can be increased by integrating synthetic augmented data to further optimize the training phase. Moreover, spatial and channel-based attention mechanisms can also be explored by emphasizing some strategic locations of the apples, which makes the recognition more accurate.
AB - Apples are one of the most consumed fruits, and they require efficient harvesting proce-dures to remains in optimal states for a longer period, especially during transportation. Therefore, automation has been adopted by many orchard operators to help in the harvesting process, which includes apple localization on the trees. The de facto sensor that is currently used for this task is the standard camera, which can capture wide view information of various apple trees from a reasonable distance. Therefore, this paper aims to produce the output mask of the apple locations on the tree automatically by using a deep semantic segmentation network. The network must be robust enough to overcome all challenges of shadow, surrounding illumination, size variations, and occlusion to produce accurate pixel-wise localization of the apples. A high-resolution deep architecture is em-bedded with an optimized design of group and shuffle operators (GSO) to produce the best apple segmentation network. GSO allows the network to reduce the dependency on a few sets of dominant convolutional filters by forcing each smaller group to contribute effectively to the task of extracting optimal apple features. The experimental results show that the proposed network, GSHR-Net, with two sets of group convolution applied to all layers produced the best mean intersection over union of 0.8045. The performance has been benchmarked with 11 other state-of-the-art deep semantic segmentation networks. For future work, the network performance can be increased by integrating synthetic augmented data to further optimize the training phase. Moreover, spatial and channel-based attention mechanisms can also be explored by emphasizing some strategic locations of the apples, which makes the recognition more accurate.
KW - apples recognition
KW - convolutional neural networks
KW - deep learning
KW - semantic segmentation
UR - http://www.scopus.com/inward/record.url?scp=85131561328&partnerID=8YFLogxK
U2 - 10.3390/agriculture12060756
DO - 10.3390/agriculture12060756
M3 - Article
AN - SCOPUS:85131561328
SN - 2077-0472
VL - 12
JO - Agriculture (Switzerland)
JF - Agriculture (Switzerland)
IS - 6
M1 - 756
ER -