Publications

2021

1. 3D shape sensing and deep learning-based segmentation of strawberries

   Justin Le Louëdec, and Grzegorz Cielniak

   Computers and Electronics in Agriculture 2021

   Abs arXiv

   Automation and robotisation of the agricultural sector are seen as a viable solution to socio-economic challenges faced by this industry. This technology often relies on intelligent perception systems pro- viding information about crops, plants and the entire environment. The challenges faced by traditional 2D vision systems can be addressed by modern 3D vision systems which enable straightforward local- isation of objects, size and shape estimation, or handling of occlusions. So far, the use of 3D sensing was mainly limited to indoor or structured environments. In this paper, we evaluate modern sensing technologies including stereo and time-of-flight cameras for 3D perception of shape in agriculture and study their usability for segmenting out soft fruit from background based on their shape. To that end, we propose a novel 3D deep neural network which exploits the organised nature of information origi- nating from the camera-based 3D sensors. We demonstrate the superior performance and efficiency of the proposed architecture compared to the state-of-the-art 3D networks. Through a simulated study, we also show the potential of the 3D sensing paradigm for object segmentation in agriculture and pro- vide insights and analysis of what shape quality is needed and expected for further analysis of crops. The results of this work should encourage researchers and companies to develop more accurate and robust 3D sensing technologies to assure their wider adoption in practical agricultural applications.

2. Gaussian map predictions for 3D surface feature localisation and counting

   Justin Le Louëdec, and Grzegorz Cielniak

   BMVC 2021 2021

   Abs arXiv

   In this paper, we propose to employ a Gaussian map representation to estimate precise location and count of 3D surface features, addressing the limitations of state-of-the-art methods based on density estimation which struggle in presence of local disturbances. Gaussian maps indicate probable object location and can be generated directly from keypoint annotations avoiding laborious and costly per-pixel annotations. We apply this method to the 3D spheroidal class of objects which can be projected into 2D shape representation enabling efficient processing by a neural network GNet, an improved UNet architecture, which generates the likely locations of surface features and their precise count. We demonstrate a practical use of this technique for counting strawberry achenes which is used as a fruit quality measure in phenotyping applications. The results of training the proposed system on several hundreds of 3D scans of strawberries from a publicly available dataset demonstrate the accuracy and precision of the system which outperforms the state-of-the-art density-based methods for this application.

2020

1. Segmentation and detection from organised 3D point clouds: A case study in broccoli head detection

   Justin Le Louëdec, Hector A Montes, Tom Duckett, and Grzegorz Cielniak

   In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops 2020

   Abs

   Autonomous harvesting is becoming an important chal- lenge and necessity in agriculture, because of the lack of labour and the growth of population needing to be fed. Per- ception is a key aspect of autonomous harvesting and is very challenging due to difficult lighting conditions, limited sens- ing technologies, occlusions, plant growth, etc. 3D vision approaches can bring several benefits addressing the afore- mentioned challenges such as localisation, size estimation, occlusion handling and shape analysis. In this paper, we propose a novel approach using 3D information for de- tecting broccoli heads based on Convolutional Neural Net- works (CNNs), exploiting the organised nature of the point clouds originating from the RGBD sensors. The proposed algorithm, tested on real-world datasets, achieves better performances than the state-of-the-art, with better accuracy and generalisation in unseen scenarios, whilst significantly reducing inference time, making it better suited for real- time in-field applications.

2. Evaluation of 3d vision systems for detection of small objects in agricultural environments

   Justin Le Louëdec, Bo Li, Grzegorz Cielniak, and  others

   2020

   Abs

   3D information provides unique information about shape, localisation and relations between objects, not found in standard 2D images. This information would be very beneficial in a large number of applications in agri- culture such as fruit picking, yield monitoring, forecasting and phenotyping. In this paper, we conducted a study on the application of modern 3D sensing technology together with the state-of-the-art machine learning algorithms for segmentation and detection of strawberries growing in real farms. We evaluate the performance of two state-of-the-art 3D sensing technologies and showcase the differences between 2D and 3D networks trained on the images and point clouds of strawberry plants and fruit. Our study highlights limitations of the current 3D vision systems for detection of small objects in outdoor applications and sets out foundations for future work on 3D perception for challenging outdoor applications such as agriculture.

3. Real-time detection of broccoli crops in 3D point clouds for autonomous robotic harvesting

   Hector A Montes, Justin Le Louëdec, Grzegorz Cielniak, and Tom Duckett

   In 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2020

   Abs

   Real-time 3D perception of the environment is crucial for the adoption and deployment of reliable autonomous harvesting robots in agriculture. Using data collected with RGB-D cameras under farm field conditions, we present two methods for processing 3D data that reliably detect mature broccoli heads. The proposed systems are efficient and enable real-time detection on depth data of broccoli crops using the organised structure of the point clouds delivered by a depth sensor. The systems are tested with datasets of two broccoli varieties collected in planted fields from two different countries. Our evaluation shows the new methods outperform state-of-the- art approaches for broccoli detection based on both 2D vision- based segmentation techniques and depth clustering using the Euclidean proximity of neighbouring points. The results show the systems are capable of accurately detecting the 3D locations of broccoli heads relative to the vehicle at high frame rates.

4. Determining shape of strawberry crops with spherical harmonics

   Justin Le Louëdec, and Grzegorz Cielniak

   In 3rd UK-RAS Conference for PhD Students and Early Career Researchers:" Robots into the real world 2020

   Abs

   Shape descriptor and shape reconstruction are two challenges found in computer vision and graphics as well as in perception for robotics, especially for some fields such as agri- robotics (robotics for agriculture). Being able to offer a reliable description of shape that can also translate directly into an high fidelity model of the shape, would be of high importance for a lot of applications such as phenotyping or agronomy. In this paper we report on our work on using spherical harmonics to offer efficient representation of strawberry shapes and we validate them by reconstructing the fruits. The reconstruction achieve extremely close results to the original shape (less than 1% deviation) and the representation reduce the complexity and improve compactness by a large factor (minimum 100).

2019

1. Deep learning investigation for chess player attention prediction using eye-tracking and game data

   Justin Le Louëdec, Thomas Guntz, James L Crowley, and Dominique Vaufreydaz

   In Proceedings of the 11th ACM Symposium on Eye Tracking Research & Applications 2019

   Abs arXiv

   This article reports on an investigation of the use of convolutional neural networks to predict the visual attention of chess players. The visual attention model described in this article has been created to generate saliency maps that capture hierarchical and spatial features of chessboard, in order to predict the probability fixation for individual pixels Using a skip-layer architecture of an autoencoder, with a unified decoder, we are able to use multiscale features to predict saliency of part of the board at different scales, showing multiple relations between pieces. We have used scan path and fixation data from players engaged in solving chess problems, to compute 6600 saliency maps associated to the corresponding chess piece configurations. This corpus is completed with synthetically generated data from actual games gathered from an online chess platform. Experiments realized using both scan-paths from chess players and the CAT2000 saliency dataset of natural images, highlights several results. Deep features, pretrained on natural images, were found to be helpful in training visual attention prediction for chess. The proposed neural network architecture is able to generate meaningful saliency maps on unseen chess configurations with good scores on standard metrics. This work provides a baseline for future work on visual attention prediction in similar contexts.