Image analysis – human style Utility companies rely increasingly on GPS-tagged images to keep track of their infrastructure. As an asset management tool, precise imagery is vital, because it lets specialists and trained repair crews evaluate where precisely to use resources on repairs – and perhaps even more valuable, to pinpoint where not to spend […]
Utility companies rely increasingly on GPS-tagged images to keep track of their infrastructure. As an asset management tool, precise imagery is vital, because it lets specialists and trained repair crews evaluate where precisely to use resources on repairs – and perhaps even more valuable, to pinpoint where not to spend resources.
As a result, utility companies often need to manage and analyse thousands of images. It’s a time-consuming task, especially if you have, say, 250,000 photos covering hundreds of miles of infrastructure, power lines, substations and assorted hardware to look at.
Scopito offers image analysis as one of our core services. This means that we can sift through thousands of images and annotate maybe a few dozen images showing critical anomalies to the drone inspection company or the utility company, drastically cutting down the time needed to let pricey specialists look at the images and decide what to do.
We have our own team of analysts working on all our inspection projects. The fact that we employ actual human beings to do the task has a couple of advantages:
First, although the ultimate dream is to automate everything via AI, we’re not quite there yet. A trained person is actually much more precise than any computer – at least until we have amassed enough data to train reliable AI algorithms.
Second is a question of flexibility. An algorithm can only do what it’s been told to do: Look for dark spots on isolators, perhaps, indicative of equipment in danger of failing. But the utility company may then decide it needs to pinpoint something else on the same batch of images, maybe buildings under the power lines.An algorithm would then have to be written and tested before this could be done. A human analyst can just be told to look for buildings and mark all images with buildings in them. It’s the work of minutes, literally, to change the search parameters this way.
But, of course, we are working towards stronger, faster automation, and all information gathered in all our projects helps us to learn more and to better train our analysts – and in the future, to form the basis of true AI image analysis.
Most projects are based on a briefing from our customer: This is what we need to find, and this type of defect typically looks like this. We then brief our team of analysts and let them get to work on the batch of images. And it really is our team: We employ our own dedicated staff of analysts.
So no crowd-sourcing here; just trained analysts. And our solution is fully scalable. In large projects, we’ll just allocate more people to the project.
Currently we are working with leading AI analytic specialists on our projects – but we are always on the lookout for other partners capable of integrating algorithms within their areas of expertise. Contact us and find out more!