Machine Vision System for Early-stage Apple Flowers and Flower Clusters Detection for Precision Thinning and Pollination

• URL: http://arxiv.org/abs/2304.09351v1
• Date: Wed, 19 Apr 2023 00:16:42 GMT
• Title: Machine Vision System for Early-stage Apple Flowers and Flower Clusters Detection for Precision Thinning and Pollination
• Authors: Salik Ram Khanal, Ranjan Sapkota, Dawood Ahmed, Uddhav Bhattarai, Manoj Karkee
• Abstract summary: We propose a vision system that detects early-stage flowers in an unstructured orchard environment using YOLOv5 object detection algorithm. The accuracy of the opened and unopened flower detection is achieved up to mAP of 81.9% in commercial orchard images.
• Score: 0.6299766708197884
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Early-stage identification of fruit flowers that are in both opened and unopened condition in an orchard environment is significant information to perform crop load management operations such as flower thinning and pollination using automated and robotic platforms. These operations are important in tree-fruit agriculture to enhance fruit quality, manage crop load, and enhance the overall profit. The recent development in agricultural automation suggests that this can be done using robotics which includes machine vision technology. In this article, we proposed a vision system that detects early-stage flowers in an unstructured orchard environment using YOLOv5 object detection algorithm. For the robotics implementation, the position of a cluster of the flower blossom is important to navigate the robot and the end effector. The centroid of individual flowers (both open and unopen) was identified and associated with flower clusters via K-means clustering. The accuracy of the opened and unopened flower detection is achieved up to mAP of 81.9% in commercial orchard images.

Related papers

• AgRegNet: A Deep Regression Network for Flower and Fruit Density Estimation, Localization, and Counting in Orchards [3.0988265713549428]
  This article proposes a deep regression-based network, AgRegNet, to estimate density, count, and location of flower and fruit in tree fruit canopies. Inspired by popular U-Net architecture, AgRegNet is a U-shaped network with an encoder-to-decoder skip connection and modified ConvNeXt-T as an encoder feature extractor.
  arXiv  Detail & Related papers  (2024-09-25T22:19:32Z)
• Generating Diverse Agricultural Data for Vision-Based Farming Applications [74.79409721178489]
  This model is capable of simulating distinct growth stages of plants, diverse soil conditions, and randomized field arrangements under varying lighting conditions. Our dataset includes 12,000 images with semantic labels, offering a comprehensive resource for computer vision tasks in precision agriculture.
  arXiv  Detail & Related papers  (2024-03-27T08:42:47Z)
• Fusion-Driven Tree Reconstruction and Fruit Localization: Advancing Precision in Agriculture [2.338903291171288]
  This study introduces an innovative methodology that harnesses the synergy of RGB imagery, LiDAR, and IMU data, to achieve intricate tree reconstructions. Experiments have been carried out in both a controlled environment and an actual peach orchard.
  arXiv  Detail & Related papers  (2023-10-23T17:44:59Z)
• AMaizeD: An End to End Pipeline for Automatic Maize Disease Detection [0.0]
  AMaizeD is an automated framework for early detection of diseases in maize crops using multispectral imagery obtained from drones. The proposed framework employs a combination of convolutional neural networks (CNNs) as feature extractors and segmentation techniques to identify both the maize plants and their associated diseases.
  arXiv  Detail & Related papers  (2023-07-23T19:58:40Z)
• Segmentation and Tracking of Vegetable Plants by Exploiting Vegetable Shape Feature for Precision Spray of Agricultural Robots [8.328453345461478]
  A novel method of Multiple Object Tracking and Tracking (MOTS) is proposed for instance segmentation and tracking of multiple vegetable plants. The proposed method is able to re-identify objects which have gone out of the camera field of view and re-appear again using the proposed data association strategy. Although the method is tested on lettuce farm, it can be applied to other similar vegetables such as broccoli and canola.
  arXiv  Detail & Related papers  (2023-06-23T14:35:17Z)
• Fruit Ripeness Classification: a Survey [59.11160990637616]
  Many automatic methods have been proposed that employ a variety of feature descriptors for the food item to be graded. Machine learning and deep learning techniques dominate the top-performing methods. Deep learning can operate on raw data and thus relieve the users from having to compute complex engineered features.
  arXiv  Detail & Related papers  (2022-12-29T19:32:20Z)
• Semantic Image Segmentation with Deep Learning for Vine Leaf Phenotyping [59.0626764544669]
  In this study, we use Deep Learning methods to semantically segment grapevine leaves images in order to develop an automated object detection system for leaf phenotyping. Our work contributes to plant lifecycle monitoring through which dynamic traits such as growth and development can be captured and quantified.
  arXiv  Detail & Related papers  (2022-10-24T14:37:09Z)
• Geometry-Aware Fruit Grasping Estimation for Robotic Harvesting in Orchards [6.963582954232132]
  geometry-aware network, A3N, is proposed to perform end-to-end instance segmentation and grasping estimation. We implement a global-to-local scanning strategy, which enables robots to accurately recognise and retrieve fruits in field environments. Overall, the robotic system achieves success rate of harvesting ranging from 70% - 85% in field harvesting experiments.
  arXiv  Detail & Related papers  (2021-12-08T16:17:26Z)
• Potato Crop Stress Identification in Aerial Images using Deep Learning-based Object Detection [60.83360138070649]
  The paper presents an approach for analyzing aerial images of a potato crop using deep neural networks. The main objective is to demonstrate automated spatial recognition of a healthy versus stressed crop at a plant level. Experimental validation demonstrated the ability for distinguishing healthy and stressed plants in field images, achieving an average Dice coefficient of 0.74.
  arXiv  Detail & Related papers  (2021-06-14T21:57:40Z)
• One-Shot Learning with Triplet Loss for Vegetation Classification Tasks [45.82374977939355]
  Triplet loss function is one of the options that can significantly improve the accuracy of the One-shot Learning tasks. Starting from 2015, many projects use Siamese networks and this kind of loss for face recognition and object classification.
  arXiv  Detail & Related papers  (2020-12-14T10:44:22Z)
• Estimating Crop Primary Productivity with Sentinel-2 and Landsat 8 using Machine Learning Methods Trained with Radiative Transfer Simulations [58.17039841385472]
  We take advantage of all parallel developments in mechanistic modeling and satellite data availability for advanced monitoring of crop productivity. Our model successfully estimates gross primary productivity across a variety of C3 crop types and environmental conditions even though it does not use any local information from the corresponding sites. This highlights its potential to map crop productivity from new satellite sensors at a global scale with the help of current Earth observation cloud computing platforms.
  arXiv  Detail & Related papers  (2020-12-07T16:23:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.