An improved cascade R-CNN and RGB-D camera-based method for dynamic cotton top bud recognition and localization in the field

A novel method for maize leaf disease classification using the RGB-D post-segmentation image data

“…The leaf region contributed more to the segmented images, that is, the extracted classification features were mainly from the leaf. Consequently, this reduced the impact on performance when the model was applied to different plots or different years ( Song et al., 2022 ).…”

A novel method for maize leaf disease classification using the RGB-D post-segmentation image data

“…The leaf region contributed more to the segmented images, that is, the extracted classification features were mainly from the leaf. Consequently, this reduced the impact on performance when the model was applied to different plots or different years ( Song et al., 2022 ).…”

Autonomous navigation method based on RGB‐D camera for a crop phenotyping robot

“…This study introduces a methodology utilizing an existing mobile robot platform designed for outdoor agricultural scenarios. The platform, featuring four-wheel independent drive and steering, measures 1.8 m in length and 1.2 m in width, rendering it apt for outdoor crop phenotyping across various agricultural environments (Song et al, 2022). The PP-LiteSeg semantic segmentation model was chosen for its capability to accurately predict crop rows in real-time.…”

A Multimodal UAV-IoT Sensing Framework for Intelligent Pest Density Estimation in Smart Agricultural Systems

“…A large body of research has employed convolutional neural networks (CNN) and their variants to automatically identify and localize pests in trap images or crop leaf images [ 20 ]. For example, methods based on two-stage detection frameworks, such as Faster R-CNN, are capable of achieving high-precision pest detection [ 21 , 22 ], while single-stage detectors represented by the YOLO series provide a favorable trade-off between real-time performance and detection accuracy [ 23 , 24 ]. In addition, some studies have incorporated fine-grained classification networks to improve discrimination among different pest species [ 25 ].…”

A novel method for maize leaf disease classification using the RGB-D post-segmentation image data

“…The leaf region contributed more to the segmented images, that is, the extracted classification features were mainly from the leaf. Consequently, this reduced the impact on performance when the model was applied to different plots or different years ( Song et al., 2022 ).…”

Autonomous navigation method based on RGB‐D camera for a crop phenotyping robot

“…This study introduces a methodology utilizing an existing mobile robot platform designed for outdoor agricultural scenarios. The platform, featuring four-wheel independent drive and steering, measures 1.8 m in length and 1.2 m in width, rendering it apt for outdoor crop phenotyping across various agricultural environments (Song et al, 2022). The PP-LiteSeg semantic segmentation model was chosen for its capability to accurately predict crop rows in real-time.…”

A Multimodal UAV-IoT Sensing Framework for Intelligent Pest Density Estimation in Smart Agricultural Systems

“…A large body of research has employed convolutional neural networks (CNN) and their variants to automatically identify and localize pests in trap images or crop leaf images [ 20 ]. For example, methods based on two-stage detection frameworks, such as Faster R-CNN, are capable of achieving high-precision pest detection [ 21 , 22 ], while single-stage detectors represented by the YOLO series provide a favorable trade-off between real-time performance and detection accuracy [ 23 , 24 ]. In addition, some studies have incorporated fine-grained classification networks to improve discrimination among different pest species [ 25 ].…”

A novel method for maize leaf disease classification using the RGB-D post-segmentation image data

“…The leaf region contributed more to the segmented images, that is, the extracted classification features were mainly from the leaf. Consequently, this reduced the impact on performance when the model was applied to different plots or different years ( Song et al., 2022 ).…”

Autonomous navigation method based on RGB‐D camera for a crop phenotyping robot

“…This study introduces a methodology utilizing an existing mobile robot platform designed for outdoor agricultural scenarios. The platform, featuring four-wheel independent drive and steering, measures 1.8 m in length and 1.2 m in width, rendering it apt for outdoor crop phenotyping across various agricultural environments (Song et al, 2022). The PP-LiteSeg semantic segmentation model was chosen for its capability to accurately predict crop rows in real-time.…”

A Multimodal UAV-IoT Sensing Framework for Intelligent Pest Density Estimation in Smart Agricultural Systems

“…A large body of research has employed convolutional neural networks (CNN) and their variants to automatically identify and localize pests in trap images or crop leaf images [ 20 ]. For example, methods based on two-stage detection frameworks, such as Faster R-CNN, are capable of achieving high-precision pest detection [ 21 , 22 ], while single-stage detectors represented by the YOLO series provide a favorable trade-off between real-time performance and detection accuracy [ 23 , 24 ]. In addition, some studies have incorporated fine-grained classification networks to improve discrimination among different pest species [ 25 ].…”