Biosystems Engineering

The data set contains raw activity data of individually housed sows in their late gestation and early lactation phase. The data is organised as trial runs and pens as it was recorded by stationary radar sensors. Metadata is available from Sauen.csv. The data is accompanied by an R-Script which produces the diagrams used for Figure 3 and 4 of the manuscript "Parturition detection in sows as test case for measuring activity behaviour in farm animals by means of radar sensors."

This dataset includes measurements collected during a greenhouse trial in 2009-2010 in Bleiswijk, The Netherlands, and results of simulations aimed at reproducing the greenhouse as a mathematical model. The greenhouse trial was conducted between October 2009 and June 2010 by WUR Greenhouse Horticulture and Flower Bulbs in Bleiswijk, the Netherlands. The experiment compared the growth of cocktail tomatoes (variety Sunstream) in greenhouse compartments with 4 different installations of artificial lighting: (1) full LED toplights, (2) full HPS (high-pressure sodium) toplights, (3) hybrid HPS toplights and LED interlights (lamps between the crop rows), (4) hybrid LED toplights and LED interlights. In all installations, lamps provided in total 170 µmol/m2/s PAR light. In the hybrid installations 50% of the artificial light came from the toplights and 50% from the interlights. This dataset includes only the data used in Katzin et al (2020): outdoor weather, indoor climate, and greenhouse climate control actions, for installations (1) and (2) only (full LED toplights and full HPS toplights), between 19 October 2009 at 15:15 and 8 February 2010 at 15:15. Model simulations were performed by GreenLight, an open-source greenhouse simulation model. Four simulations were performed: (1) A simulation of the indoor climate with HPS lamps; (2) A simulation of the indoor climate with LED lamps; (3) A simulation of the energy use with HPS lamps ; (4) A simulation of the energy use with LED lamps.

This dataset includes measurements collected during a greenhouse trial in 2009-2010 in Bleiswijk, The Netherlands, and results of simulations aimed at reproducing the greenhouse as a mathematical model. The greenhouse trial was conducted between October 2009 and June 2010 by WUR Greenhouse Horticulture and Flower Bulbs in Bleiswijk, the Netherlands. The experiment compared the growth of cocktail tomatoes (variety Sunstream) in greenhouse compartments with 4 different installations of artificial lighting: (1) full LED toplights, (2) full HPS (high-pressure sodium) toplights, (3) hybrid HPS toplights and LED interlights (lamps between the crop rows), (4) hybrid LED toplights and LED interlights. In all installations, lamps provided in total 170 µmol/m2/s PAR light. In the hybrid installations 50% of the artificial light came from the toplights and 50% from the interlights. This dataset includes only the data used in Katzin et al (2020): outdoor weather, indoor climate, and greenhouse climate control actions, for installations (1) and (2) only (full LED toplights and full HPS toplights), between 19 October 2009 at 15:15 and 8 February 2010 at 15:15. Model simulations were performed by GreenLight, an open-source greenhouse simulation model. Four simulations were performed: (1) A simulation of the indoor climate with HPS lamps; (2) A simulation of the indoor climate with LED lamps; (3) A simulation of the energy use with HPS lamps ; (4) A simulation of the energy use with LED lamps.

This dataset serves as supplementary material for the research paper titled 'Sim2real flower detection towards automated Calendula harvesting', which was published in the October 2023 issue of Biosystems Engineering. Within this upload, you will find a collection of both authentic and computer-generated images featuring Calendula (Calendula officinalis L.) flowers. Additionally, we have included the resources and original data utilized in generating the synthetic images. This dataset proves instrumental in demonstrating the successful transference of a deep neural network from simulation to real-world applications. The contents of this upload includes: Original RGB and depth images of Calendula flowers captured in a natural flower field, complete with bounding box annotations. The test data employed in our experiments. Unedited RGB images that were used in the photogrammetry pipeline. Three-dimensional models representing Calendula flowers. The resulting dataset of synthetic images. The synthetic dataset follows the Synthetic Optimized Labeled Objects (SOLO) Dataset Schema, as defined in the Unity Perception package. For completeness, the data scheme is also included in the upload. For more comprehensive details regarding this dataset and its associated metadata, we invite you to consult the published article in Biosystems Engineering, available at the following link: https://doi.org/10.1016/j.biosystemseng.2023.08.016.

This publicly available dataset contains 1613 RGB-D images of field-grown broccoli plants. The dataset also includes the polygon and circle annotations of the broccoli heads. The broccoli heads in the images were subject to various degrees of natural and man-made leaf occlusion. The images were acquired in July and August 2020 on a broccoli field in Sexbierum (The Netherlands). The broccoli cultivar was Ironman. The dataset belongs to the paper “Image-based size estimation of broccoli heads under varying degrees of occlusion”. This paper has been published at Biosystems Engineering journal: https://doi.org/10.1016/j.biosystemseng.2021.06.001

A zip-file with all raw data collected with the vibrometer. asc-files with the collated vibrometer data as described in the paper. In addition, records of the mass of the onions and images of cut bulbs. The data are organised according to cultivar name as presented in Table 1 of the paper.

A zip-file with all raw data collected with the vibrometer. asc-files with the collated vibrometer data as described in the paper. In addition, records of the mass of the onions and images of cut bulbs. The data are organised according to cultivar name as presented in Table 1 of the paper.

A zip-file with all raw data collected with the vibrometer. asc-files with the collated vibrometer data as described in the paper. In addition, records of the mass of the onions and images of cut bulbs. The data are organised according to cultivar name as presented in Table 1 of the paper.

Detection of Pig Parts (csv, coco, tfrecord and voc formats) The annotated dataset is provided with different formats for easy deployment: this include, coco, tfrecord, csv and voc. Furthermore, an augmented version of the dataset is provided with many more annotated images. The augmented dataset (recommended) is also provided with annotations using the above formats. The collected image dataset was annotated by two trained individuals with an animal behaviour background. It encompassed a variety of scenarios, for example, pigs in close contact with one another and under various lighting conditions. We configured a set of pre-processing stages to augment the dataset, applying arbitrary scaling and horizontal flipping. We also manipulated the colour of the pixels and randomly altered the brightness and contrast using the hue, saturation, value (HSV) colour space. The detection dataset comprised a total of 51193 instances (26533 AFBI + 24660 AUF) across 2781 images (1556 AFBI + 1225 AUF); each pig within an image was manually annotated into two parts: head and rear. A bounding box1 was applied manually on the head and rear of all pigs in a pen. The bounding box denotes the location and size of each pig part. Contact Between Pigs (csv format) An additional dataset was annotated to validate the interaction method, i.e., the processing stage that feeds from the detection method. This dataset consisted of images from both farms used in this framework. The total number of images of this dataset was 670 images; with sets of 376 and 294 images to represent AFBI and AUF datasets, respectively. A similar procedure was followed for selecting the image samples to diversify the dataset. This new dataset was annotated by an animal behaviour scientist who scanned all images to score interactions using a predefined ethogram. Any contact between one pig head and another pig rear was scored in a csv file. The entirety of the dataset comprised four classes (per image) as the following: no contact, 1 contact; 2 contacts; 3 or more contacts. Very few images (of the AUF dataset) included more than three contacts; therefore, these were combined in one class to achieve a more balanced class distribution. Code Code for the detection method, i.e., detecting heads and rears of individual pigs. Code for the interaction method, using the above detector to index interactions between head and rear of pigs.

Detection of Pig Parts (csv, coco, tfrecord and voc formats) The annotated dataset is provided with different formats for easy deployment: this include, coco, tfrecord, csv and voc. Furthermore, an augmented version of the dataset is provided with many more annotated images. The augmented dataset (recommended) is also provided with annotations using the above formats. The collected image dataset was annotated by two trained individuals with an animal behaviour background. It encompassed a variety of scenarios, for example, pigs in close contact with one another and under various lighting conditions. We configured a set of pre-processing stages to augment the dataset, applying arbitrary scaling and horizontal flipping. We also manipulated the colour of the pixels and randomly altered the brightness and contrast using the hue, saturation, value (HSV) colour space. The detection dataset comprised a total of 51193 instances (26533 AFBI + 24660 AUF) across 2781 images (1556 AFBI + 1225 AUF); each pig within an image was manually annotated into two parts: head and rear. A bounding box1 was applied manually on the head and rear of all pigs in a pen. The bounding box denotes the location and size of each pig part. Contact Between Pigs (csv format) An additional dataset was annotated to validate the interaction method, i.e., the processing stage that feeds from the detection method. This dataset consisted of images from both farms used in this framework. The total number of images of this dataset was 670 images; with sets of 376 and 294 images to represent AFBI and AUF datasets, respectively. A similar procedure was followed for selecting the image samples to diversify the dataset. This new dataset was annotated by an animal behaviour scientist who scanned all images to score interactions using a predefined ethogram. Any contact between one pig head and another pig rear was scored in a csv file. The entirety of the dataset comprised four classes (per image) as the following: no contact, 1 contact; 2 contacts; 3 or more contacts. Very few images (of the AUF dataset) included more than three contacts; therefore, these were combined in one class to achieve a more balanced class distribution. Code Code for the detection method, i.e., detecting heads and rears of individual pigs. Code for the interaction method, using the above detector to index interactions between head and rear of pigs.