n image processing computer application to automatically assess the vigor of three-day-old soybean (Glycine max [L.] Merrill.) seedlings was developed. The software operates on acquired digital images of soybean seed lots placed on a paper towel. Soybean seedlings were extracted away from the paper towel and converted into various digital representations. These representations were used to analyze the seedlings and segment them into normal and abnormal categories. The normal seedlings were further processed so that a one-pixel-wide summary structure of the shape of the seedling was produced. From this summary structure, the software classified the seedlings into six type categories based on their shape. Each normal seedling was processed to remove the cotyledon portion of the summary structure based on the type category it fell into. The remaining summary structure, with the cotyledon removed, was then used to compute the length of each seedling in pixels. From these length measurements, speed of growth and uniformity of growth values were computed. These two values were normalized and combined into a zero to 1,000 vigor index for the seed lot. Combined with the post-processing corrective features, this computer software was able to achieve highly accurate and standardized measurements of each soybean seedling, providing an alternative to the current method of manually measuring soybean seedlings for speed and uniformity of growth when performing a vigor test. A eed vigor is defined as "those seed properties which determine the potential for rapid, uniform emergence and development of normal seedlings under a wide range of field conditions" (AOSA, 1983). A seed vigor test is a series of analyses on a seed lot that provide an indication of its emergence potential in the field. While there are many different ways to measure a seed lot's vigor, one of the most important is to determine a seed lot's speed and uniformity of growth as described in the seed vigor definition. The two main limitations of performing a vigor test manually are 1) results of a vigor test may vary from laboratory to laboratory because of the subjective nature of most vigor tests and 2) many vigor tests take excessive time to acquire results. These two limitations were addressed by designing computer software that measures the seedlings represented by a digital image and computes the vigor index from those measurements. S fter being germinated for 3 days, an image of the soybean seedlings was acquired with a large scanning box that allowed the placement of two flatbed scanners side by side. The scanners were operated in serial by a controlling software application to obtain two images that were combined together to represent all 50 seedlings (Figure 1). A From this image, the soybean seedlings were segmented away from the dark brown paper towel background and converted into various digital representations. These representations included a list of pixel positions representing each seedling's silhouette (binary object) and a 1-pixel wide summary ...