Pipe feature identification using pipe inspection data analysis

1. A method, comprising:

operating a mobile pipe inspection platform to obtain sensor data for the interior of a pipe, wherein the sensor data comprises image data obtained using a camera and additional pipe inspection data obtained using one or more other sensors;

analyzing, using a processor, the sensor data using a trained model, wherein the trained model is trained using a dataset including sensor data of pipe interiors, metadata identifying pipe feature locations contained within the sensor data of the dataset and metadata classifying pipe features contained within the sensor data of the dataset;

identifying, using the processor, a pipe feature location within the sensor data;

classifying, using the processor, a pipe feature of the sensor data;

wherein the classifying comprises:

associating a classification result for the image data with a classification result for terahertz image data; and

thereafter providing a material identification of the pipe feature;

combining analysis of the additional pipe inspection data with analysis of the image data; and

thereafter producing, using the processor, an output including an indication of the pipe feature, wherein the pipe feature is one or more of: a crack in a pipe wall, erosion in a pipe wall, sediment buildup in a pipe wall, and an intrusion through a pipe wall.

2. The method of claim 1 , wherein the analyzing comprises automated object detection using the sensor data.

3. The method of claim 1 , wherein the analyzing comprising receiving user input to select a pipe feature.

4. The method of claim 1 , wherein the trained model comprises a trained neural network.

5. The method of claim 4 , wherein the trained neural network comprises a neural network trained using reference visual images with annotated pipe features or pipe feature locations.

6. The method of claim 5 , wherein the annotated pipe features are annotated by a human operator.

7. The method of claim 1 , wherein the additional pipe inspection data comprises one or more of laser pipe inspection data, sonar pipe inspection data, infrared pipe inspection data, and terahertz pipe inspection data.

8. A system, comprising:

a mobile pipe inspection platform that comprises one or more sensors to obtain sensor data for the interior of a pipe, wherein the sensor data comprises image data obtained using a camera and additional pipe inspection data obtained using one or more other sensors;

a computer system operatively coupled to the mobile pipe inspection platform and configured to:

analyze the sensor data using a trained model, wherein the trained model is trained using a dataset including sensor data of pipe interiors, metadata identifying pipe feature locations contained within the sensor data of the dataset and metadata classifying pipe features contained within the sensor data of the dataset;

identify a pipe feature location within the sensor data; and

classify a pipe feature of the sensor data;

wherein to classify comprises:

associating a classification result for the image data with a classification result for terahertz image data; and

thereafter providing a material identification of the pipe feature;

combine analysis of the additional pipe inspection data with analysis of the image data; and

thereafter produce an output including an indication of the pipe feature, wherein the pipe feature is one or more of: a crack in a pipe wall, erosion in a pipe wall, sediment buildup in a pipe wall, and an intrusion through a pipe wall.

9. The system of claim 8 , wherein the analysis of the sensor data comprises automated object identification using the sensor data.

10. The system of claim 8 , wherein the analysis of the sensor data comprises receiving user input to select a pipe feature.

11. The system of claim 8 , wherein the trained model comprises a trained neural network.

12. The system of claim 11 , wherein the trained neural network comprises a neural network trained using reference visual images with annotated pipe features or pipe feature locations.

13. The system of claim 12 , wherein the annotated pipe features are annotated by a human operator.

14. The system of claim 8 , wherein the additional pipe inspection data comprises one or more of laser pipe inspection data, sonar pipe inspection data, infrared pipe inspection data, and terahertz pipe inspection data.

15. A product, comprising:

a non-transitory storage device that stores code that is executable by a processor, the code comprising:

code that analyzes sensor data obtained from a mobile pipe inspection platform, wherein the sensor data comprises image data obtained using a camera and additional pipe inspection data obtained using one or more other sensors, using a trained model, where the trained model is trained using a dataset including sensor data of pipe interiors, metadata identifying pipe feature locations contained within the sensor data of the dataset and metadata classifying pipe features contained within the sensor data of the dataset;

code that performs:

identifying a pipe feature location within the sensor data; and

classifying a pipe feature of the sensor data;

wherein the classifying comprises:

associating a classification result for the image data with a classification result for terahertz image data; and

thereafter providing a material identification of the pipe feature;

code that combines analysis of the additional pipe inspection data with analysis of the image data; and

code that thereafter produces an output including an indication of the pipe feature, wherein the pipe feature is one or more of: a crack in a pipe wall, erosion in a pipe wall, sediment buildup in a pipe wall, and an intrusion through a pipe wall.