Register for free to
view the full innovation profile.
ITAD uses thermal imaging to identify loose rock in hanging walls of hard rock mines which can directly afterwards be verified by a ML supported acoustic classification model
About
The ITAD is a lightweight handheld device that early entry examiners can use to identify loose rock in the hanging wall. Thereby, preventing involuntary fall-of-ground (FOG) leading to injury or fatalities. The device improves the examiner’s ability to identify loose rock by providing visual processing to increase the field-of-vision, and machine-learning acoustic sensing. The visual processing develops a panoramic image of the thermal image to increase the field of vision. The acoustic sensing uses a trained machine learning algorithm to identify the sound of loose rock after it has been tapped. The ITAD has a rugged design to withstand harsh mining environments. It is a single-handed device with color-coded buttons and a touch screen for user inputs. ITAD consists of two subsystems: the thermal imaging and the acoustic sensing systems. These systems are integrated into the device to determine the integrity of hanging walls near the rock face. When interacting with the thermal imaging system, the user will be engaging with a stream from the thermal camera. Once an area of interest has been identified, an image of the thermal stream will be captured, and manipulated using a panoramic imaging algorithm to increase the field of vision of a captured thermal image, allowing the user to see a larger image of the rock being investigated. South African mines are notorious for being physical constrained without space to see much of the hanging wall. These imaging algorithms allow the user to increase the field-of-vision for the thermal imager to investigate larger portions of the rock. The acoustic system allows the user to record the audio response of the rock, after it being sounded by the user. Thereafter, the audio sample gets parsed through a machine learning model to classify whether the rock is deemed ‘safe’ or ‘unsafe’. For further development and training of the ML model, an opportunity is provided for the user to input whether they think it is safe or unsafe.
Key Benefits
It enhances hard rock gold and platinum mine re-entry inspectors' ability to identify loose rocks from the hanging wall through thermal imaging. It eliminates the need to touch the roof during this step of determining whether the rock is loose. It gives a AI-generated classification on whether tapped rock from the hanging wall is safe or unsafe. It mitigates human error during the re-entry inspection in hard rock mines. It can speed up the duration of the re-entry inspection. A specialist is not needed to do the re-entry inspection.
Applications
The Integrated Thermal Acoustic Device (ITAD) assists miners in detecting loose rock using a non-contact thermal camera method to analyse and display suspected loose rock to the user. This potentially speeds up the process of identifying loose rock. Thereafter, the assessment can be further verified using an acoustic method of tapping the rock with a pinch bar and analysing the acoustic response. This process is known as ‘sounding’ the rock. However, the acoustic method can be used on its own too.