Back to Top
Skip to main content

Twitter Icon Linkedin Icon Facebook Icon Instagram Icon You Tube Icon Flickr Icon

Smart Methane Emission Detection System Development
Project Number
Last Reviewed Dated

The goal of this project is to develop an autonomous, real-time methane leak detection technology, the Smart Methane Emission Detection System, which applies machine learning techniques to passive optical sensing modalities to mitigate emissions through early detection. Phase 1 is leveraging previous SwRI research and experience to develop the prototype methane detection system with integrated optical sensors and the embedded processing unit. Phase 2 will focus on integration and field-testing of the prototype system along with demonstration to the DOE within a representative controlled environment. The system will target the following key features:

Table 1. Key Features of the Smart Methane Emission Detection System



Low False Alarm Rates

Less than 0.5% (number of events incorrectly classified as leaks).

Autonomous Detection

No need for a human to be in the loop – the system acquires, processes, and makes autonomous decisions on whether or not a hazardous substance was observed, using machine learning algorithms.

Near Real-Time Detection

The time between acquiring data and obtaining an output from the system is only a few minutes.

Non-Intrusive, Passive Technology

No need to retrofit existing equipment and facilities. The proposed technology is passive in nature, thus eliminating safety and operational restrictions.

Target Platform

Ability to deploy technology in a stationary platform for monitoring facilities such as refineries and pump stations.


Ability to nimbly integrate new detection techniques into system for other types of target substances.


Southwest Research Institute (SwRI), San Antonio, TX 78238
Falcon Inspection (camera in-kind contributor)
IRCameras (camera in-kind contributor)


While much has been accomplished in developing optical sensors for imaging methane leaks, limited work has been accomplished in developing methane emission detection technologies that meet three key critical criteria for effective methane emission mitigation:

  1. Autonomy (no need for a human to be in the loop)
  2. High reliability (low false alarm rates)
  3. Real-time performance (ability to detect emissions within a few minutes once it starts)

This project focuses on these three key aspects and significantly advances the state-of-the-art in methane emission detection using optically-based sensing technologies.


An autonomous, real-time methane leak detection system facilitates early detection of emissions before they become a larger problem. Compressor station operators will be able to identify failing equipment in aging infrastructure and replace faulty components expediently, resulting in methane emissions being reduced significantly through early detection of non-compliant equipment. This project is expected to produce the following outcomes and/or impacts:

  • Develop a system to reliably, accurately and autonomously identify methane leaks at critical midstream sections of the natural gas distribution network in real-time for the purpose of mitigating methane emissions.
  • Add a high degree of automation to the process of methane leak detection to minimize sources of human error, minimize response time to a leak event, and maximize midstream visibility.
  • Assist in the quantification process by providing a means of collecting temporal and spatial image data of a leak event.
  • Reduce operational costs of emissions detection technologies by significantly minimizing the need for operator involvement.Provide a solution that is scalable, cost-effective, and non-intrusive.
  • Reduce methane emissions by early real-time, autonomous detection of methane leaks.
Accomplishments (most recent listed first)
  • The SLED/M system went through initial testing, and successfully detected methane leaks from the live Mid-Wave Infrared (MWIR) camera stream in real-time.
  • The embedded detection algorithm (running on a NVIDIA Tegra TX1 chip) was successfully integrated with the real-time MWIR camera stream.
  • The algorithm passed all initial laboratory validation tests with target performance measures.
  • The embedded system was verified to drive the cameras at high framerate and execute the detection algorithm in real-time (~1Hz).
  • Sensor (cameras) setup, configuration, and calibration was completed.
  • Initial test data were labeled for classifier training. Initial convolutional neural network (CNN) was configured for training.
  • The following equipment for the budget period was ordered and received: MWIR camera from IRCameras, MWIR camera from Falcon Inspection, thermal camera (SwRI in-kind).
  • Tests and data acquisition for budget period 1 are underway.
  • Feature extraction and analysis tasks were completed.
  • Development of initial leaks and non-leak events detection algorithm was completed.
Current Status

Camera setup, configuration, and calibration was completed. The Mid-Wave Infrared (MWIR) cameras are being utilized as the transducers to capture frames in search of methane gases. The cameras are configured and calibrated to effectively capture the environment under varying illumination, weather conditions, and distances. An initial set of tests was performed under representative conditions to establish a baseline database containing methane leaks of various concentrations, distances, and scenarios. These conditions include an initial set of varying ambient temperature conditions, cloud cover, presence, lack of obstacles (such as piping), and varying wind (including stagnant) conditions. Subsequent tests were performed with more representative scenarios of methane leaks as well as leaks of other types of gases that might be present at compression station-type facilities that could trigger false alarms. This was done to increase the accuracy and robustness of the algorithms. An initial set of test data was labeled for classifier training. An initial convolutional neural network (CNN) was also configured for training.

Development of the classifier progressed through many iterations of data collection via fine tuning of the algorithm’s architecture, structure, and parameters. Extensive data has been collected covering a range of lighting, background, and other ambient conditions. This data is needed in order to provide a sufficient variety of training and validation data to guarantee a robust algorithm.

The algorithm has successfully detected live methane leaks under multiple operating conditions, and is currently undergoing rigorous testing to establish performance metrics. The embedded software, which drives the MWIR camera, executes the detection algorithm to detect methane emissions, and reports these events automatically through a live HTTP stream. The stream has the capability of displaying both live video feed from the camera and a detection overlay, which indicates when and where the system has detected methane, as shown in Figure 1. The software is implemented to run onboard an NVIDIA Tegra TX1 board. It is capable of taking the camera video stream and producing methane detection predictions in real-time (~1 Hz). An initial hardware design has been completed and partially implemented for testing the full system in relevant conditions. The hardware houses the camera and all necessary computational components necessary to perform the methane detections and reporting.

Figure 1: SLED/M Methane Detection (Left, Highlighted in Red)
Figure 1: SLED/M Methane Detection (Left, Highlighted in Red)


Project Start
Project End
DOE Contribution


Performer Contribution


Contact Information

NETL – Joseph B. Renk III ( or 412-386-6406)
SwRI – Maria S. Araujo ( or 210-522-3730)

Additional Information