Image Credit Oliver Cameron (https://news.voyage.auto/an-introduction-to-lidar-the-key-self-driving-car-sensor-a7e405590cff) “A Picture is worth thousand words”, this proverb refers to the notion that a complex idea can be conveyed with just a single picture, and the picture conveys its meaning or essence more effectively than a the words does.
However, in digital age, this proverb has become more relevant than ever, as today images are really able to convey much more and capture much more information than any other form of data. This has been made possible by availability of compute capacity, and storage capabilities that are unprecedented.
Every industry has some challenge areas where the applicability of image analytics as effective solution has been proved. Be it Lifesciences industry, where deep learning algorithms are able to do complete medical diagnostics by analyzing images. Industry leading players like Zebra Medical Vision are transforming the medical image diagnosis industry by incorporating AI and Deep Learning to reduce human intervention and provide accurate diagnosis results .
Similarly, Manufacturing and High Tech industry has been relying on image analytics for years for quality check of manufactured goods, and reduce defects. There are mature solutions from multiple vendors that support range of problems in defect identification one such solution is WebSPECTOR for surface inspection .
Images analytics has not only enabled adding value as a discreet solution, but also it integrates with other evolving technologies to enable an end to end solution that solves business problem. Improvement in drone technology has made it convenient and practical to take pictures of areas that were previously not possible or where extensive effort was required. When such drone imagery combines with intelligent asset monitoring solutions, it helps in monitoring those assets that were previously unconnected. Utility industry has grappled with this problem of large number of linear assets spread across extended geographic regions.
Managing such vast expense of assets and monitoring those was an expensive affair for the industry. As an example, oil transmission industry spends more than USD 36 Bn every year on maintenance and monitoring of pipelines. Further, various monitoring solutions have the problem of scale, that is, in case we have to increase the frequency of such inspections, the cost of such maintenance operations becomes prohibitive, and hence leading to huge open window where assets are not monitored.
This exposes organization to risk of leakage and possible regulatory non-compliance. However, in these scenarios drone technologies combined with advanced image analytics can help in avoiding the field visits to monitor these assets. The images captured can be processed with advanced image analytics algorithms that can recognize various assets and identify any damages that requires attention. GIS solutions providers like Harris Technologies are developing deep learning based solutions to automate asset monitoring for Power and Transmission industry to improve the compliance and reduce response time.
The applicability of these solutions is not only limited to distributed assets, but also in retail stores many retailers want to capture the intelligence of shop floor inventory. For retail industry any stock out of inventory in store shelf leads to lost sale and hence impact on per square feet revenue. Hence, industry is making use of image analytics in many ways. Images from cameras can identify if the shelves need to be restocked. Also this data can help in understanding the relation between a particular merchandise placement and the interest it generates for customers.
Organizations like Coca cola are leveraging image technologies for shelf management and inventory placement to maximize sales, optimize inventory and make supply chain more efficient. Industry leading products like Trax retail are helping these organizations solve the problem that leads to huge sales opportunity loss due to stock outs. As Digital transformation has impacted every industry from Automotive to Manufacturing, image analytics is creating tangible value and use cases. The application areas of this confluence of technologies, (AI, Machine Learning, Deep Learning, & Image Analysis) is only limited by our imagination.
The original article can be found at: