Abstract
An early estimation of the exact number of fruits, flowers, and trees helps farmers to make better decisions on cultivation practices, plant disease prevention, and the size of harvest labor force. The current practice of yield estimation based on manual counting of fruits or flowers by workers is a time consuming and expensive process and it is not feasible for large fields. Automatic yield estimation based on robotic agriculture provides a viable solution in this regard. In a typical image classification process, the task is not only to specify the presence or absence of a given object on a specific location, while counting how many objects are present in the scene. The success of these tasks largely depends on the availability of a large amount of training samples. This paper presents a detector of bunches of one fruit, grape, based on a deep convolutional neural network trained to detect vine bunches directly on the field. Experimental results show a 91% mean Average Precision.
Highlights
Precision agriculture evaluates spatial and temporal variability of field data through automatic collection and digitization of extensive information databases
While in other works the authors try used the Intersection over Union (IoU) measure (Equation (1)), which allows us to estito limit those variations as much as possible, on the contrary, we have tried to include mate the precision in the overlap between a bounding box obtained by the classifier and as many variations as possible in our training using dataset augmentation, so that the that defined as ground truth that is the one hand drawn during the ‘labelling’ process
To evaluate the effectiveness of the proposed approach for bunches detection, we used the Intersection over Union (IoU) measure (Equation (1)), which allows us to estimate the precision in the overlap between a bounding box obtained by the classifier and that defined as ground truth that is the one hand drawn during the ‘labelling’ process. 11 of 21
Summary
Precision agriculture evaluates spatial and temporal variability of field data through automatic collection and digitization of extensive information databases. Different types of sensors are applied to develop high-efficiency approaches to optimize input use, maximize crop production, reduce wastes, guarantee environmental sustainability, and obtain economic benefits [1,2,3,4]. These specific approaches apply to viticulture in terms of efficient use of inputs, such as fertilizers, water, chemicals, or organic products [5,6]. Vineyards are characterized by high spatial and temporal heterogeneity and are influenced by pedo-morphological characteristics, climate, phenology, and cropping practices [12] These variables can influence grape yields and quality, and their prediction is the main goal of precision viticulture. Farmers can be encouraged to pursue the economic benefits and achieve the desired oenological results by the latest technologies combined with decision support techniques [13,14,15]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
