6 min read
Mining has been a part of human history for centuries. It is essential to some regions of the world as their primary export product, creating employment opportunities and establishing related businesses in nearby communities.
Mining continues to be one of the essential industries, alongside agriculture, energy, and many manufacturing sectors. In 2020, the annual GDP contribution from the mining and quarrying (except oil and gas) industry was US 56 billion dollars while providing close to 550,000 job opportunities. Technological advances have made mines increasingly more productive, environmentally friendly, and safer. Industry 4.0 has also ushered in IIoT sensor technologies to monitor mining operations remotely and equipment in real-time, further reducing the need to deploy humans.
Why 3D machine vision is optimal for the mining industry
3D machine vision is a field of industrial scanning that utilizes three methods to digitize target objects or scenes: interferometry, time of flight, or laser triangulation. Known for its high speed and fast scanning rate, laser triangulation is the most robust method that supports a wide range of industrial activities.
Laser triangulation is achieved by illuminating the target object with a pencil or a fanned laser beam. The light bounces off the object and is seen by an off-axis sensor (imager). Where the reflected light is detected within the sensor will correspond to the object’s distance (depth) from the laser projector. Here is an illustration depicting a typical laser triangulation scanner setup. Most importantly, laser triangulation is independent of ambient lighting and does not require a high degree of contrast, making it ideal in dark environments.
About the Small Automation, Big Impacts series
The automation journey series is based on Hermary’s extensive field experience and many proofs of concepts we have worked on with our partners and end customers. We want to inspire the industry and help automation experts by sharing what 3D machine vision can do for them. As a whole, we will improve industrial automation together.
We want to hear about your journeys with industrial automation. Share your story with us, and we will make sure that it is published on our website for the whole industry to see.
Data captured by 3D machine vision is stored as point clouds. The data points are a list of three-dimensional values representing the target object’s surface. Depending on the application, industrial scanners can scan up to 812,500 points per second, creating high-resolution 3D representations in seconds. There are also no constraints on the size of the object – scanners can be daisy-chained, and data points can be stitched to create a complete scene. Furthermore, data points can be filtered using proper algorithms to reduce noise, ensuring downstream processes have pristine data input.
Collecting volumetric data in mines with 3D machine vision
Automating the segmentation process in mining, for example, coal or iron ore pellets, can save processing time and improve product quality. A reliable classification process often depends on accurate volumetric information of the particles. 2D machine vision misses the crucial depth measurement and often uses image intensity changes to approximate the missing value. This method is slow, prone to error, and often results in inaccurate sortation.
3D machine vision creates point cloud data that consists of thousands of X, Y, and Z coordinates. A series of these coordinates translate to a real-world object’s width, length, and height (depth) down to millimeters, capturing the necessary information to calculate the volumetric data. Most notably, systems using 3D triangulation data produce deterministic results, guaranteeing the segmentation process’ accuracy and improving product quality.
As variations in ore pellet size negatively impact a furnace’s melting efficiency, it is in an iron ore manufacturer’s best interest to perfect the segmentation process. A 2008 research published in Minerals Engineering illustrates that when 3D triangulation is used in sorting ore pellets, the system can distinguish size class with only a 0.5 mm difference. By deploying a 3D scanning system, iron miners can significantly improve their product quality and thus customer satisfaction.
System engineers have been trying to overcome visual occlusion caused by mineral deposits overlapping or touching each other on the conveyor belt. The sawmill industry has used multi-scanner systems to eliminate visual occlusion to improve measurement accuracies. In such cases, one 3D scanner can be mounted at an angle on either side of the conveyor belt. The data from both scanners can be stitched and filtered to calculate the volume or even density. Both pieces of information are valuable for further classification or analysis, which, with improved accuracy, directly boosts a mining company’s bottom line.
Small Automation, Big Impacts is a collection of industrial automation journeys made possible by 3D machine vision.
We believe sharing these stories is a way to stay connected with our partners and positively impact the manufacturing community.
This free eBook also includes a bonus chapter to help automation professionals easily decide when to use 2D and 3D machine vision.
All this is just one click away in our new eBook, Small Automation, Big Impacts.
Laser triangulation is unaffected by lighting or the object’s contrast against the background. 3D machine vision, therefore, is unmatched when scanning dark or black minerals, such as coal, cobalt, or ilmenite (titanium oxide), on black conveyor belts.
An additional mining application that can be improved by 3D scanning is equipment monitoring. A conveyor system supported by idlers is widely used to carry ores from the deposit site to a storage area. Idlers near the boring machine are subjected to more wear as the falling ores constantly impact them. Worn idlers can cause the belt to misalign or its movement sluggish, slowing down the entire operation. Using 3D machine vision to monitor the belt alignment is a proactive way to give the control room data needed to schedule predictive maintenance.
The benefits of monitoring the belt are twofold. Ideally, the ore should be deposited onto the center of the conveyor belt for maximum efficiency. Offset loading of the ore (meaning the ore load is concentrated on one side) can cause uneven wear, belt misalignment, and unnecessary spillages. Monitoring the belt movement can alert operators of potential misalignment and adjust the offloading chute before costly mistakes happen.
The mining industry is one of the oldest industries in the world. It has been around for thousands of years. But it’s also an industry that has seen continuous technological advancements that made the industry safer, more efficient, and environmentally sustainable with each passing decade.
In a 2020 research published by Mining, machine vision data successfully segmented mixed material images on a conveyor belt using convolutional neural networks. A convolutional neural network is a type of artificial intelligence that can be trained to detect visual features in images. These findings will hopefully inspire more applications using 3D machine vision and point cloud data to achieve 100% quality assurance.
According to Deloitte’s Tracking the trends 2021 report on the mining industry, improving overall safety should be every company’s top priority. In aiming to achieve zero harm, mining companies establish trust with their workers, communities, and stakeholders. The report also urges companies to accelerate technology and automation adoption that builds on workplace wellness.
Machine vision’s application is still vastly underutilized in the mining industry. 3D machine vision is especially suitable for underground shafts as it utilizes lasers as the illumination source. Empowering robots with 3D machine vision (vision-guided robots) may be used to replace dangerous or laborious tasks, further moving workers out of harm’s way in mines. 3D machine vision can provide crucial insights into the health of mining equipment. Should an incident occur, data analysis may help reveal the cause, which can be used to construct preventative measures.
Industry: Mining (excluding oil & gas)
- Reduces the need for deploying humans
- Empowers the control room with data for predictive maintenance
- Accurate dimensional information in seconds
- Improved segmentation process
- Fewer manual workers needed
- Alerts operators of unbalanced load
- Upskilling workers to perform value-added tasks
- 3D machine vision can scan dark deposits on black conveyor belts
- High-speed scanners can accommodate even faster conveyor speeds
- Use of deep learning to identify mixed materials
- Vision-guided robots to help with mining operations
- Increased product quality
- Reduced reliance on manual work
- Eliminate unnecessary spillage
- Minimize equipment failure