A team including FSU’s Oliver Steinbock and FAMU’s Beni Dangi has trained a machine learning algorithm to identify the chemical composition of various salts. They recorded 7,500 photos of 42 different types of salt stains and translated each image into parameters that capture features such as deposit area, compactness, and texture. The images were then translated into numbers that encode the patterns’ arrangement of crystals.
“We were surprised at how well this worked,” Steinbock says. “Who would think that from a photo, you can tell the difference between sodium chloride and potassium chloride?” The research has many potential applications, such as analyzing samples collected by a rover during planetary exploration, testing materials for lab safety, or low-cost blood analysis.
View Related Expert Profiles: Go to Source
