ICON: AI intelligent colony counting opens up a new experimental method

Food safety is one of the most concerning issues for consumers, and the total bacterial count is an important criterion for determining the quality of food, drugs, and other products. Colony counting is widely used in microbial testing, quality supervision, and environmental monitoring in fields such as water, food safety, and medicine.

Chocolate is a food made from cocoa beans, and there are three main methods of making it:Products that directly use cocoa butter or use cocoa butter as the main raw material; Products made from edible oils or cocoa butter as raw materials; The finished product is made from cocoa beans as the main raw material and water as the solvent through processes such as grinding, alkali dissolution, and filtration. However, there is a high risk of microbial contamination in the processed raw materials. The use of an automatic colony counter can quickly count the microbial colonies in the raw materials, ensuring the safety and hygiene level of the raw materials.

During the chocolate production process, the quality inspection department mainly detects the total number of bacterial colonies in both raw materials and finished products. As the raw materials for chocolate production need to be ground, impurities may be mixed in during sample preparation and inoculation, resulting in poor homogenization of the finished product. Some particles may exist in the form of impurities in the petri dish due to incomplete dissolution, and bubbles may also appear on the surface of the petri dish.

Traditional artificial colony counting is prone to misjudgment of bubbles, impurities, and other factors, leading to biased results and affecting judgment standards. In addition, it requires colony counting of large quantities of samples, which undoubtedly becomes a major challenge for laboratory bacterial counting.ICONThe automatic colony counter adopts industrial grade high-definition technologyCMOSThe camera has a minimum recognition ability for bacterial colonies0.05mmThrough image acquisition and recognition, it solves the problem of manual counting and accurately counts the entire dish.

ICON adoptsAIIntelligent algorithms, which differ from traditional algorithms in terms of recognition singularity, can learn and train unfamiliar colony morphology, greatly improving the recognition rate of colonies. They can effectively filter out impurities and bubbles that interfere with counting. With just one click counting, the whole dish counting can be quickly completed, reducing human errors and greatly improving experimental efficiency, fundamentally solving the pain points of the existing colony counting mode in the food industry.