A image dataset for indoor soybean investigation

Breeding good varieties is an important guarantee for improving quality and yield of soybeans. Phenotypic information, such as plant height, number of main stem nodes and pod number per plant, is the key factor of evaluating soybean varieties. An accurate acquisition of the aforementioned information is an important goal of soybean investigation. Traditional methods of manual measurement and counting are time-consuming and laborious, and the data error inevitably occurs. The method of obtaining soybean investigation information based on machine vision can effectively reduce human workload and labor intensity, but it heavily relies on the result of artificial depoding and placement. Furthermore, there is a bottleneck in extracting phenotypic information of plants, which is far from the one-stop high-throughput acquisition of key seed information. Therefore, the construction of the image dataset of indoor soybean is very important for carrying out the research on a method of high-throughput and accurate acquisition of soybean plant phenotype information and further realizing automatic and intelligent soybean investigation. The present dataset consists of three parts: original image data, annotation file data and test image data, covering in-vitro pods and plants of typical soybean varieties, in which the former includes non-overlapping in-vitro pods and overlapping ones, and the latter includes plants of single branch, double branches and complex branches. The annotation file data includes instance segmentation labeling of main stems in soybean plants, detection box labeling of main stem nodes, detection box labeling of in-body pods, and the labeling of instance segmentation and detection box of the in-vitro pods. Moreover, the horizontal and rotation boxes are employed regarding the detection box labeling of soybean plant pods and in-vitro pods, with the total amount of 5.6GB. This dataset can provide valuable basic image data resources for in-vitro pod detection, in-body pod detection and plant morphological analysis of object detection and semantic segmentation models, as well as important value for promoting the research of intelligent soybean investigation by means of offering one-stop high-throughput acquisition of soybean key information.