Apple flower detection using deep convolutional networks

Abstract

To optimize fruit production, a portion of the flowers and fruitlets of apple trees must be removed early in the growing season. The proportion to be removed is determined by the bloom intensity, i.e., the number of flowers present in the orchard. Several automated computer vision systems have been proposed to estimate bloom intensity, but their overall performance is still far from satisfactory even in relatively controlled environments. With the goal of devising a technique for flower identification which is robust to clutter and to changes in illumination, this paper presents a method in which a pre-trained convolutional neural network is finetuned to become specially sensitive to flowers. Experimental results on a challenging dataset demonstrate that our method significantly outperforms three approaches that represent the state of the art in flower detection, with recall and precision rates higher than 90%. Moreover, a performance assessment on three additional datasets previously unseen by the network, which consist of different flower species and were acquired under different conditions, reveals that the proposed method highly surpasses baseline approaches in terms of generalization capability. 1 , 2 1 The citation information for this article is: P. A. Dias, A. Tabb, and H. Medeiros, Apple flower detection using deep convolutional networks,

Improvements to and large‐scale evaluation of a robotic kiwifruit harvester

“…Soft computing approaches have become the de facto approach for agricultural robotics as they have been found to be able to handle dynamic conditions reliably (Y. Huang et al, ). They have been utilised for a wide range of agricultural tasks: harvesting, yield estimation, weed‐spraying, pollination, and crop management (Bargoti & Underwood, a, b; Dias, Tabb, & Medeiros, ; Kurosaki et al, ; Nachtigall, Araujo, & Nachtigall, ; Sa et al, ; Wan, Toudeshki, Tan, & Ehsani, ; Wang, Song, & He, ; Zhang et al, ). Detection of apples (Bargoti & Underwood, b; Dias et al, ; Inthiyaz, Kishore, & Madhav, ; Moallem, Serajoddin, & Pourghassem, ; Prasad et al, ; Puttemans, Vanbrabant, Tits, & Goedemé, ; Soleimani Pour, Chegini, Zarafshan, & Massah, ) and strawberries (Habaragamuwa et al, ; Puttemans et al, ) has shown good results with detection rates up to 90% of the fruit under real‐world orchard conditions.…”

“…They have been utilised for a wide range of agricultural tasks: harvesting, yield estimation, weed-spraying, pollination, and crop management (Bargoti & Underwood, 2017a, 2017bDias, Tabb, & Medeiros, 2018;Kurosaki et al, 2011;Nachtigall, Araujo, & Nachtigall, 2016;Sa et al, 2016;Wan, Toudeshki, Tan, & Ehsani, 2018;Wang, Song, & He, 2017;Zhang et al, 2017). Detection of apples (Bargoti & Underwood, 2017b;Dias et al, 2018;Inthiyaz, Kishore, & Madhav, 2018;Moallem, Serajoddin, & Pourghassem, 2017;Prasad et al, 2018;Puttemans, Vanbrabant, Tits, & Goedemé, 2017;Soleimani Pour, Chegini, Zarafshan, & Massah, 2018) and strawberries (Habaragamuwa et al, 2018;Puttemans et al, 2017) has shown good results with detection rates up to 90% of the fruit under real-world orchard conditions. A kiwifruit detection system using semantic segmentation was able to detect 76.0% of kiwifruit in a real-world orchard (H. A. .…”

Autonomous pollination of individual kiwifruit flowers: Toward a robotic kiwifruit pollinator

“…Soft computing techniques have been utilized for a wide range of harvesting, yield estimation, weed-spraying, pollination, and crop management within orchards (Bargoti & Underwood, 2017a, 2017bDias, Tabb, & Medeiros, 2018;Kurosaki et al, 2011;Nachtigall, Araujo, & Nachtigall, 2016;Sa et al, 2016;Wang, Song, & He, 2017;Wan, Toudeshki, Tan, & Ehsani, 2018;Zhang et al, 2017). Detection of Apples (Bargoti & Underwood, 2017b;Dias et al, 2018;Inthiyaz, Kishore, & Madhav, 2018;Moallem, Serajoddin, & Pourghassem, 2017;Prasad et al, 2018;Puttemans, Vanbrabant, Tits, & Goedemé, 2017;Soleimani Pour, Chegini, Zarafshan, & Massah, 2018) and strawberries (Habaragamuwa et al, 2018;Puttemans et al, 2017) have shown good results with detection rates up to 90% of the fruit under real-world orchard conditions. A kiwifruit detection system using semantic segmentation was able to detect 76.0% of kiwifruit in a real-world orchard (Williams et al, 2018), showing promise for the detection of the flowers under similar conditions.…”

Multispecies Fruit Flower Detection Using a Refined Semantic Segmentation Network

“…Our previous work in [11] introduced a novel approach for apple flower detection that relies on a fine-tuned Clarifai CNN [18] to classify individual superpixels composing an image. That method highly outperformed color-based approaches, especially in terms of generalization to datasets composed of different flower species and acquired in uncontrolled environments.…”