Abstract
Latest advances in artificial intelligence, particularly in object recognition and segmentation, provide unprecedented opportunities for precision agriculture. This work investigates the use of state-of-the-art AI models, namely Meta’s Segment Anything (SAM) and GroundingDino , for the task of grape cluster detection in vineyards. Three different methods aimed at enhancing the instance segmentation process are proposed: (i) SAM-Refine (SAM-R), which refines a previously proposed depth-based clustering approach, referred to as DepthSeg, using SAM; (ii) SAM-Segmentation (SAM-S), which integrates SAM with a pre-trained semantic segmentation model to improve cluster separation; and (iii) AutoSAM-Dino (ASD), which eliminates the need for manual labeling and transfer learning through the combined use of GroundingDino and SAM. Analysis is conducted on both the object counting and pixel-level segmentation accuracy against a manually labeled ground truth. Metrics such as mean Average Precision (mAP), Intersection over Union (IoU), and precision and recall are calculated to assess the system performance. Compared to the original DepthSeg algorithm, SAM-R slightly advances object counting (mAP: +0.5%) and excels in pixel-level segmentation (IoU: +17.0%). SAM-S, despite a mAP decrease, improves segmentation accuracy (IoU: +13.9%, Precision: +9.2%, Recall: +11.7%). Similarly, ASD, although with a lower mAP, shows significant accuracy enhancement (IoU: +7.8%, Precision: +4.2%, Recall: +4.9%). Additionally, from a labor effort point of view, instance segmentation techniques require much less time for training than manual labeling. • Novel AI approaches for grape detection: SAM-Refine, SAM-Segmentation, AutoSAM-Dino. • SAM-Refine improves segmentation accuracy (IoU +17.0%) vs. DepthSeg. • AutoSAM-Dino removes manual labeling, enabling efficient deployment. • Depth-based methods robust under suboptimal lighting conditions. • Proposed methods reduce labor while keeping high segmentation quality.
Abstract from the paper’s OpenAlex record; open copy licensed cc-by. Read at the source ↗
The text the model read, with the source record’s inline markup (italics, sub- and superscripts) rendered and its character entities decoded; quote offsets refer to the raw text.
Claims 13
Claim · model-written summary
SAM-R improved object counting mAP by 0.5% over DepthSeg.
Verbatim
e0“Compared to the original DepthSeg algorithm, SAM-R slightly advances object counting (mAP: +0.5%)”
t0“grape bunch”
e1“Three different methods aimed at enhancing the instance segmentation process are proposed: (i) SAM-Refine (SAM-R), which refines a previously proposed depth-based clustering approach, referred to as DepthSeg, using SAM; (ii) SAM-Segmentation (SAM-S), which integrates SAM with a pre-trained semantic segmentation model to improve cluster separation; and (iii) AutoSAM-Dino (ASD), which eliminates the need for manual labeling and transfer learning through the combined use of GroundingDino and SAM.”
Claim · model-written summary
SAM-R improved pixel-level segmentation IoU by 17.0% over DepthSeg.
Verbatim
e0“SAM-R slightly advances object counting (mAP: +0.5%) and excels in pixel-level segmentation (IoU: +17.0%)”
t0“grape bunch”
e1“Three different methods aimed at enhancing the instance segmentation process are proposed: (i) SAM-Refine (SAM-R), which refines a previously proposed depth-based clustering approach, referred to as DepthSeg, using SAM; (ii) SAM-Segmentation (SAM-S), which integrates SAM with a pre-trained semantic segmentation model to improve cluster separation; and (iii) AutoSAM-Dino (ASD), which eliminates the need for manual labeling and transfer learning through the combined use of GroundingDino and SAM.”
Claim · model-written summary
SAM-S had decreased mAP.
Verbatim
e0“SAM-S, despite a mAP decrease”
t0“grape bunch”
e1“Three different methods aimed at enhancing the instance segmentation process are proposed: (i) SAM-Refine (SAM-R), which refines a previously proposed depth-based clustering approach, referred to as DepthSeg, using SAM; (ii) SAM-Segmentation (SAM-S), which integrates SAM with a pre-trained semantic segmentation model to improve cluster separation; and (iii) AutoSAM-Dino (ASD), which eliminates the need for manual labeling and transfer learning through the combined use of GroundingDino and SAM.”
Claim · model-written summary
SAM-S improved segmentation IoU by 13.9%.
Verbatim
e0“SAM-S, despite a mAP decrease, improves segmentation accuracy (IoU: +13.9%, Precision: +9.2%, Recall: +11.7%)”
t0“grape bunch”
e1“Three different methods aimed at enhancing the instance segmentation process are proposed: (i) SAM-Refine (SAM-R), which refines a previously proposed depth-based clustering approach, referred to as DepthSeg, using SAM; (ii) SAM-Segmentation (SAM-S), which integrates SAM with a pre-trained semantic segmentation model to improve cluster separation; and (iii) AutoSAM-Dino (ASD), which eliminates the need for manual labeling and transfer learning through the combined use of GroundingDino and SAM.”
Claim · model-written summary
SAM-S improved segmentation precision by 9.2%.
Verbatim
e0“improves segmentation accuracy (IoU: +13.9%, Precision: +9.2%, Recall: +11.7%)”
t0“grape bunch”
e1“Three different methods aimed at enhancing the instance segmentation process are proposed: (i) SAM-Refine (SAM-R), which refines a previously proposed depth-based clustering approach, referred to as DepthSeg, using SAM; (ii) SAM-Segmentation (SAM-S), which integrates SAM with a pre-trained semantic segmentation model to improve cluster separation; and (iii) AutoSAM-Dino (ASD), which eliminates the need for manual labeling and transfer learning through the combined use of GroundingDino and SAM.”
Claim · model-written summary
SAM-S improved segmentation recall by 11.7%.
Verbatim
e0“improves segmentation accuracy (IoU: +13.9%, Precision: +9.2%, Recall: +11.7%)”
t0“grape bunch”
e1“Three different methods aimed at enhancing the instance segmentation process are proposed: (i) SAM-Refine (SAM-R), which refines a previously proposed depth-based clustering approach, referred to as DepthSeg, using SAM; (ii) SAM-Segmentation (SAM-S), which integrates SAM with a pre-trained semantic segmentation model to improve cluster separation; and (iii) AutoSAM-Dino (ASD), which eliminates the need for manual labeling and transfer learning through the combined use of GroundingDino and SAM.”
Claim · model-written summary
ASD had lower mAP.
Verbatim
e0“ASD, although with a lower mAP”
t0“grape bunch”
e1“Three different methods aimed at enhancing the instance segmentation process are proposed: (i) SAM-Refine (SAM-R), which refines a previously proposed depth-based clustering approach, referred to as DepthSeg, using SAM; (ii) SAM-Segmentation (SAM-S), which integrates SAM with a pre-trained semantic segmentation model to improve cluster separation; and (iii) AutoSAM-Dino (ASD), which eliminates the need for manual labeling and transfer learning through the combined use of GroundingDino and SAM.”
Claim · model-written summary
ASD improved IoU by 7.8%.
Verbatim
e0“ASD, although with a lower mAP, shows significant accuracy enhancement (IoU: +7.8%, Precision: +4.2%, Recall: +4.9%)”
t0“grape bunch”
e1“Three different methods aimed at enhancing the instance segmentation process are proposed: (i) SAM-Refine (SAM-R), which refines a previously proposed depth-based clustering approach, referred to as DepthSeg, using SAM; (ii) SAM-Segmentation (SAM-S), which integrates SAM with a pre-trained semantic segmentation model to improve cluster separation; and (iii) AutoSAM-Dino (ASD), which eliminates the need for manual labeling and transfer learning through the combined use of GroundingDino and SAM.”
Claim · model-written summary
ASD improved precision by 4.2%.
Verbatim
e0“significant accuracy enhancement (IoU: +7.8%, Precision: +4.2%, Recall: +4.9%)”
t0“grape bunch”
e1“Three different methods aimed at enhancing the instance segmentation process are proposed: (i) SAM-Refine (SAM-R), which refines a previously proposed depth-based clustering approach, referred to as DepthSeg, using SAM; (ii) SAM-Segmentation (SAM-S), which integrates SAM with a pre-trained semantic segmentation model to improve cluster separation; and (iii) AutoSAM-Dino (ASD), which eliminates the need for manual labeling and transfer learning through the combined use of GroundingDino and SAM.”
Claim · model-written summary
ASD improved recall by 4.9%.
Verbatim
e0“significant accuracy enhancement (IoU: +7.8%, Precision: +4.2%, Recall: +4.9%)”
t0“grape bunch”
e1“Three different methods aimed at enhancing the instance segmentation process are proposed: (i) SAM-Refine (SAM-R), which refines a previously proposed depth-based clustering approach, referred to as DepthSeg, using SAM; (ii) SAM-Segmentation (SAM-S), which integrates SAM with a pre-trained semantic segmentation model to improve cluster separation; and (iii) AutoSAM-Dino (ASD), which eliminates the need for manual labeling and transfer learning through the combined use of GroundingDino and SAM.”
Claim · model-written summary
Instance segmentation techniques required much less training time than manual labeling.
Verbatim
e0“instance segmentation techniques require much less time for training than manual labeling”
t0“grape bunch”
e1“Three different methods aimed at enhancing the instance segmentation process are proposed: (i) SAM-Refine (SAM-R), which refines a previously proposed depth-based clustering approach, referred to as DepthSeg, using SAM; (ii) SAM-Segmentation (SAM-S), which integrates SAM with a pre-trained semantic segmentation model to improve cluster separation; and (iii) AutoSAM-Dino (ASD), which eliminates the need for manual labeling and transfer learning through the combined use of GroundingDino and SAM.”
Claim · model-written summary
AutoSAM-Dino eliminated manual labeling and transfer learning.
Verbatim
e0“AutoSAM-Dino (ASD), which eliminates the need for manual labeling and transfer learning”
t0“grape bunch”
e1“Three different methods aimed at enhancing the instance segmentation process are proposed: (i) SAM-Refine (SAM-R), which refines a previously proposed depth-based clustering approach, referred to as DepthSeg, using SAM; (ii) SAM-Segmentation (SAM-S), which integrates SAM with a pre-trained semantic segmentation model to improve cluster separation; and (iii) AutoSAM-Dino (ASD), which eliminates the need for manual labeling and transfer learning through the combined use of GroundingDino and SAM.”
Claim · model-written summary
Depth-based methods were robust under suboptimal lighting.
Not reportedComparatorValue
Verbatim
e0“Depth-based methods robust under suboptimal lighting conditions”
t0“grape bunch”
e1“Three different methods aimed at enhancing the instance segmentation process are proposed: (i) SAM-Refine (SAM-R), which refines a previously proposed depth-based clustering approach, referred to as DepthSeg, using SAM; (ii) SAM-Segmentation (SAM-S), which integrates SAM with a pre-trained semantic segmentation model to improve cluster separation; and (iii) AutoSAM-Dino (ASD), which eliminates the need for manual labeling and transfer learning through the combined use of GroundingDino and SAM.”