Using computer vision and machine learning, we are able to track features and dimensions of moving objects in real time. The hardware that we use is extremely affordable and lightweight, while still providing the quality needed to make appropriate business decisions. The software we have designed is flexible enough to work on millimeter wide saw blades or 40 foot logs. Currently, we are implementing our computer vision software to analyze the saw blade teeth of a lumber mill and determine if it needs repaired or replaced using machine learning to categorize the blade. Due to our utilization of machine learning, our accuracy in this categorization will increase over time. We hope to make this process efficient and exact, therefore increasing overall productivity of the lumber mill.

We have created a reporting software tool to present useful data to timber scaling companies in the form of tables and pivot tables. All timber data is stored in a customized database system and is retrieved by a user-friendly reporting tool we built in a Google compute engine virtual machine using Python scripts, SQL queries, and Google apps scripts. Reports are automatically emailed to the specified target as pdf documents. We report on up to several thousand logs a day, and it only takes a few minutes. We are hoping to cut that time down even more by implementing the reporting system into our front-end data management tool, thereby removing the need for the VM and additional scripts.