Abstract: A sensing device has a sensor. An edge device that has a processor and memory receives a data stream from the sensor. A communication device connects the edge device to a cloud computing system over a network for processing the data stream and for sending the processed data stream to the cloud computing system. The memory includes instructions for execution by the processor to deploy an inference engine using a pipeline. A capture module receives and stores the data stream within the memory. A pre-processing module modifies the data stream in the memory to compensate for defects contained therein, thereby forming a modified data stream. A deep learning module identifies inferences within the modified data stream, trains a model based upon the inferences, and combines the modified data stream, the model, and the inferences to form the processed data stream. A logging module stores the processed data stream.

Abstract: A cloud platform has a message broker and a web portal thereon. An edge device connects to the cloud platform and has a camera thereon, memory for storing computer readable instructions, and a processor for executing the computer readable instructions. A video stream comprising a plurality of video frames is captured from the camera. A set of coordinates to define a region of interest is generated to insert into at least one of the plurality of video frames to form a modified video stream. The modified video stream is processed with an inference module to obtain a plurality of inferences. The modified video stream and the plurality of inferences is sent to the cloud platform web portal to display output relating to the modified video stream and the plurality of inferences thereon.

Abstract: A cloud server has memory for storing computer readable instructions and a processor for executing the computer readable instruction. A pipe inspection video is received for storage in a video storage container. A plurality of frames is extracted from the pipe inspection video. A plurality of receiver containers is formed to perform artificial intelligence operations the plurality of frames in parallel. The plurality of frames is sent to the plurality of receiver containers to obtain metrics relating to the pipe inspection video. The plurality of frames is returned to the video storage container. The plurality of frames is compiled into a processed pipe inspection video.