What type of 'AI' does Cyclomedia use?
The term "AI" encompasses a wide range of technologies. Over the past decade, we have successfully integrated various Computer Vision (CV) and Machine Learning (ML) techniques into our operations. These techniques include object detection, image segmentation, and classification, which have been instrumental in enhancing our data processing capabilities.
- Object Detection: This involves identifying and locating objects within images, allowing us to extract valuable information from visual data efficiently.
- Image Segmentation: By dividing images into meaningful segments, we can analyze specific regions more effectively, improving the accuracy of our data interpretations.
- Classification: This process involves categorizing objects or elements within images into predefined classes, facilitating organized and systematic data analysis.
These technologies have been in production at Cyclomedia for many years, demonstrating our long-standing commitment to leveraging AI for operational excellence.
Individual privacy
Blurring of identifiable features
Cyclomedia uses AI to blur faces and license plates in our imagery in accordance to GDPR requirements.
Gathering of personal information
Cyclomedia does not gather any identification information from our imagery.
Data training
Data to train our internal models is gathered internally from our own data and is not sourced from copyrighted works.
Why do we use AI to perform analytics?
While one may not need AI to gather asset data, using AI does allow it to come at a lower financial cost and at a much faster rate. This saves Governments time and resources which are already limited.
FAQ
Cyclomedia's solutions help Local Governments keep their communities safe and habitable. By providing information on assets such as traffic signs, roads, dilapidated properties, ADA ramps, and more, we give organizations the information they need to make informed decisions, ultimately improving quality of life for their constituents.
Cyclomedia randomly spots our blurring AI to verify it performs according to spec, and upgrade the model if we see any issues.
Some of Cyclomedia's solutions include ADA Ramps, Manholes, Road Surface Analysis, and recognition of Dilapidated Properties. These solutions (amongst many others) each serve to improve the safety and quality of government assets and, ultimately, their communities as a whole.
Large Language Models
Looking ahead, we are exploring the potential of Large Language Models (LLMs). Although we have not yet incorporated these advanced AI models into our production systems, we recognize their transformative potential. LLMs, known for their ability to understand and generate human-like text, represent the cutting edge of AI research. By investigating their applications, we aim to stay at the forefront of AI innovation and continuously enhance our technological capabilities.