Image Dataset of Local Indonesian Soybean Seed Varieties (Anjasmoro, Grobogan, and DEGA-1)
Description
This dataset contains high-resolution scanned images of local Indonesian soybean seed varieties, namely Anjasmoro, Grobogan, and DEGA-1. The images were captured using a flatbed scanner under standardized conditions to ensure consistency in lighting, resolution, and background. Each seed was placed flat on the scanner glass to minimize distortion and maximize the visibility of morphological characteristics. The dataset includes both raw seed images and segmented seed images. The segmentation process was applied to isolate individual seeds from the background, allowing for more precise analysis of shape, size, color, and texture. These features are crucial indicators in assessing seed quality and potential productivity, and thus the dataset can serve as a valuable resource for research in seed science, plant breeding, and agricultural technology. Potential applications of this dataset include seed classification, variety identification, quality assessment, machine learning model training for automated seed recognition, and digital phenotyping studies. Researchers and practitioners in plant breeding and agricultural informatics may particularly benefit from this open dataset as it provides a benchmark for developing seed analysis methodologies.