Machine learning applied to computer vision and patternrecognition is a type of artificial intelligence that hasadvanced rapidly in the last 10 to 15 years, spurred forwardby breakthroughs in deep convolutional neural networks.These state-of-the-art methods are poised to become widelyused in environmental monitoring applications as a resultof the increasing abundance of data available from differentimaging platforms (e.g., fixed-point cameras, drone surveys,high-resolution satellite data) that can be analyzed toComputer vision and pattern recognition (CVPR) toolsadvance our ability to use imagery and camera-basedtools in cost-effective ways for environmental monitoring.Although these techniques offer great potential, somechallenges remain, such as the need for large sets of labeledimages for model training and validation and optimizedhardware and software to ensure that the models can betrained effectively and in a reasonable amount of time. Here,we have overcome many of these rate-limiting challenges byusing a diverse image library built across multiple projectscoupled with staff expertise and onsite computing resources.We are working toward a fully automated SPI processingsystem and also are moving to develop CVPR analyticaltools for other imaging platforms and data sets.observe, model, and understand environmental conditions.