TY - JOUR
T1 - Artificial Intelligence-Based Segmentation and Classification of Plant Images with Missing Parts and Fractal Dimension Estimation
AU - Batchuluun, Ganbayar
AU - Kim, Seung Gu
AU - Kim, Jung Soo
AU - Mahmood, Tahir
AU - Park, Kang Ryoung
N1 - Publisher Copyright:
© 2024 by the authors.
PY - 2024/11
Y1 - 2024/11
N2 - Existing research on image-based plant classification has demonstrated high performance using artificial intelligence algorithms. However, limited camera viewing angles can cause parts of the plant to be invisible in the acquired images, leading to an inaccurate classification. However, this issue has not been addressed by previous research. Hence, our study aims to introduce a method to improve classification performance by taking these limitations into account; specifically, we incorporated both segmentation and classification networks structured as shallow networks to expedite the processing times. The proposed shallow plant segmentation network (Shal-PSN) performs adversarial learning based on a discriminator network; and a shallow plant classification network (Shal-PCN) with applied residual connections was also implemented. Moreover, the fractal dimension estimation is used in this study for analyzing the segmentation results. Additionally, this study evaluated the performance of the proposed Shal-PSN that achieved the dice scores (DSs) of 87.43% and 85.71% with PlantVillage and open leaf image (OLID-I) open datasets, respectively, in instances where 40–60% of plant parts were missing. Moreover, the results demonstrate that the proposed method increased the classification accuracy from 41.16% to 90.51% in the same instances. Overall, our approach achieved superior performance compared to the existing state-of-the-art classification methods.
AB - Existing research on image-based plant classification has demonstrated high performance using artificial intelligence algorithms. However, limited camera viewing angles can cause parts of the plant to be invisible in the acquired images, leading to an inaccurate classification. However, this issue has not been addressed by previous research. Hence, our study aims to introduce a method to improve classification performance by taking these limitations into account; specifically, we incorporated both segmentation and classification networks structured as shallow networks to expedite the processing times. The proposed shallow plant segmentation network (Shal-PSN) performs adversarial learning based on a discriminator network; and a shallow plant classification network (Shal-PCN) with applied residual connections was also implemented. Moreover, the fractal dimension estimation is used in this study for analyzing the segmentation results. Additionally, this study evaluated the performance of the proposed Shal-PSN that achieved the dice scores (DSs) of 87.43% and 85.71% with PlantVillage and open leaf image (OLID-I) open datasets, respectively, in instances where 40–60% of plant parts were missing. Moreover, the results demonstrate that the proposed method increased the classification accuracy from 41.16% to 90.51% in the same instances. Overall, our approach achieved superior performance compared to the existing state-of-the-art classification methods.
KW - deep learning
KW - fractal dimension
KW - limited camera viewing angle
KW - missing plant parts
KW - plant image classification and segmentation
KW - plant images
UR - http://www.scopus.com/inward/record.url?scp=85210443750&partnerID=8YFLogxK
U2 - 10.3390/fractalfract8110633
DO - 10.3390/fractalfract8110633
M3 - Article
AN - SCOPUS:85210443750
SN - 2504-3110
VL - 8
JO - Fractal and Fractional
JF - Fractal and Fractional
IS - 11
M1 - 633
ER -