TSGYE: Two-Stage Grape Yield Estimation

Deng, Geng; Geng, Tianyu; He, Chengxin; Wang, Xinao; He, Bangjun; Duan, Lei

doi:10.1007/978-3-030-63820-7_66

Cited by 5 publications

(8 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A similar score of 91% was obtained on the GrapceCS-ML dataset [101]. The Yolov4 model reached 96.96% mAP on the dataset published by Santos et al [96,97]. They contributed to this dataset by labeling the individual berries.…”

Section: Performance Comparisonsupporting

confidence: 69%

“…The 3D position is used as an identifier to avoid counting one grape twice. Mask R-CNN and multiple Yolo models [39] were compared on the dataset published by Santos et al [96,97]. Yolo models were also compared on red grape detection from smartphone images [98].…”

Section: Deep Learningmentioning

confidence: 99%

“…Deep Learning [81,83,84,91,93,94,[96][97][98][99][100][101][102]105,[107][108][109][110]112,116]. 300 images, five red and white varieties, natural lighting [96].…”

Section: Key-point Detection Classification and Clustering (Svm And D...mentioning

confidence: 99%

“…This difference is explained by the structure of the models, as Mask R-CNN uses a complex structure to detect and segment objects, and by the small size of the database. Deng et al [97] used the Hough transform for circle counting after grape detection with the Yolov4 model (it is similar to the method proposed by Rudolph et al for flower counting [53]). One limitation of this algorithm is its sensibility to berries' apparent size (the radius must be known).…”

Section: Deep Learningmentioning

confidence: 99%

“…For instance, the fast radial symmetry transform studied by [125] had a recall rate varying between 69% and 89%. Similarly, the Hough Transform for berry counting proposed by Deng et al had performance varying from 1.7% MAE counting error to 24.85% depending on the variety [97]. This is due to the need for a known radius size before counting, as it can lead to worse performance when the pixel berry size is too small.…”

Section: Performance Comparisonmentioning

confidence: 99%

See 4 more Smart Citations

Computer Vision and Deep Learning for Precision Viticulture

Mohimont

Alin

Rondeau³

et al. 2022

Agronomy

View full text Add to dashboard Cite

During the last decades, researchers have developed novel computing methods to help viticulturists solve their problems, primarily those linked to yield estimation of their crops. This article aims to summarize the existing research associated with computer vision and viticulture. It focuses on approaches that use RGB images directly obtained from parcels, ranging from classic image analysis methods to Machine Learning, including novel Deep Learning techniques. We intend to produce a complete analysis accessible to everyone, including non-specialized readers, to discuss the recent progress of artificial intelligence (AI) in viticulture. To this purpose, we present work focusing on detecting grapevine flowers, grapes, and berries in the first sections of this article. In the last sections, we present different methods for yield estimation and the problems that arise with this task.

show abstract

Section: Performance Comparisonsupporting

confidence: 69%

Section: Deep Learningmentioning

confidence: 99%

“…Deep Learning [81,83,84,91,93,94,[96][97][98][99][100][101][102]105,[107][108][109][110]112,116]. 300 images, five red and white varieties, natural lighting [96].…”

Section: Key-point Detection Classification and Clustering (Svm And D...mentioning

confidence: 99%

Section: Deep Learningmentioning

confidence: 99%

Section: Performance Comparisonmentioning

confidence: 99%

See 3 more Smart Citations

Computer Vision and Deep Learning for Precision Viticulture

Mohimont

Alin

Rondeau³

et al. 2022

Agronomy

View full text Add to dashboard Cite

show abstract

An In-Field Dynamic Vision-Based Analysis for Vineyard Yield Estimation

Ahmedt-Aristizabal,

Smith,

Khokher

et al. 2024

IEEE Access

View full text Add to dashboard Cite

STEWIE: eSTimating grapE berries number and radius from images using a Weakly supervIsed nEural network

Botturi,

Gnutti,

Nuzzi

et al. 2023

2023 IEEE International Workshop on Metrology for Agriculture and Forestry (MetroAgriFor)

View full text Add to dashboard Cite

Counting tasks with overlapping and occluded targets are often tackled by means of neural networks outputting density maps. While this approach has been proven to be highly effective for crowd-counting tasks, it has not been exploited extensively in other fields (like fruit counting). Furthermore, this approach has never been used to infer the shape or the size of the recognized objects. In this paper, we present a novel deep learning-based methodology to automatically estimate the number of grape berries present in an image and evaluate their average radius as a double output of the network. For the model training, we employ a public dataset consisting of 300 vines images, where each berry center has been dot-annotated. Since the dataset does not directly provide information about the berry radii, we first develop a numerical optimization methodology to calculate the radius of the berries, by exploiting the dot annotations, some prior knowledge (berry maximum size), and a current state-of-the-art segmentation model. Then, we employ the combined information (berry center and radius) to train a custom neural network that outputs two density maps, from which we infer the number of berries in the image and their average size.

show abstract

TSGYE: Two-Stage Grape Yield Estimation

Cited by 5 publications

References 13 publications

Computer Vision and Deep Learning for Precision Viticulture

Computer Vision and Deep Learning for Precision Viticulture

An In-Field Dynamic Vision-Based Analysis for Vineyard Yield Estimation

STEWIE: eSTimating grapE berries number and radius from images using a Weakly supervIsed nEural network

Contact Info

Product

Resources

About