All our solutions are based on plant data that helps us benchmark your plant's performance, identify your bottlenecks and rank the opportunities. Through our holistic approach we evaluate the overall alarm system performance, downtime logs and correlate those with availalble process data and the control system scheme.
We utilise state of the art convolutional neural network based computer vision algorithms that are custom designed to your specific application. We offer detailed camera design, sensor fusion recommendation and edge hardware. Our CV software can detect product quality and granularity as well as dynamic events like blockages, liquefaction, contamination etc.
Through automation, Espy reduces the workload on mine operators so they can focus on higher level tasks. Alarm system performance optimisation, control loop performance improvement, high level control scheme optimisation are a few examples of the experties of the Espy team.
No one technology can eliminate contamination completely. Therefore, we propose a holistic approach, where we offer our CV system to complement other technologies and fully integrate with the existing plant infrastructure to detect contamination in stockpiles, in the excavator bucket, on truck trays or in the plant on conveyors/feeders/screens and in hoppers.
The Espy Computer Vision System has successfully been implemented to detect contaminations of timber, roof bolt, metal objects, liquids and also different product size contamination (lump in fines). Our Computer Vision System has the capability to learn over time, we can train the system on new data on demand or we can host the training environment for you on premise or in the cloud.
Our contamination detection systems end where the contamination is detected and removed from the product. As opposed to most AI companies, Espy offers end to end solutions specifically tailored to the needs of the mining industry. Our solutions do not require cloud access and platform subscriptions and reach well over the realm of AI. We integrate with any process control systems and high level ERP to maximise the benefits to our clients.
Espy combines the power of computer vision and existing instrumentation infrastructure through sensor fusion to maximise value extraction of your data.
Visual detection offers the richest and most condensed data source available, which makes it superior to most conventional ore contamination detection sensors. The Espy Computer Vision System can offer a reliable, visual based detection of any kind of contamination as long as they are at least partially visible (>100mm on a 1.5m wide conveyor).
At its core the Espy Computer Vision System relies on object detection deep neural network architectures that are capable of real-time (sub seconds) feature extraction, object localisation bounding box regression and object classification.
The inference system signals any image frame with possible timber visible. If the confidence of the recognition is high enough the system signals to the PCS, where along with mass tracking the contaminated segments are marked and tracked so a diverter gate or diversion plough can be engaged for the right amount of time to discard the contamination.
Roof bolt contamination in coal can damage conveyor belts and block processing equipment causing major downtime. Steel bolts are conventionally picked out by tramp magnets and detected by metal detectors. However, poly and fibreglass roof bolts pass through these systems.
The Espy Computer Vision System can offer a reliable, visual based detection of any kind of roof bolts as long as they at least partially visible (>100mm). In the system, a specialised machine vision camera is mounted above the conveyor where it monitors ~2m section of the conveyor at 5 frame per second processing rate.
The Espy Computer Vision System fully integrates with the local control system so it can receive feedback from the metal detector and send commands to trip a conveyor or track the contamination along the belt so it can be automatically rejected.