Originally published by EETimes
AI-based approaches are used in various applications across industries to control and optimize the operation of specific use cases. These solutions cover a wide range of applications ranging from agriculture and automobiles to healthcare and manufacturing. Real-time image and video processing is one such use case where AI plays a significant role in analyzing, identifying, and extracting the required information from an image. Using camera imagery, specialized sensors process the data to allow for further analysis. As the quality of an image or video (which is a series of images) depends on factors like the type/size of the camera sensors as well as lighting conditions, it becomes crucial to implement a versatile image processing system.
In low light conditions, fewer photons are captured by the camera sensor, making the image dim. In addition to poor lighting conditions, high dynamic range, heavy fog, rain, and sudden changes in intensity make the image quality poor and full of noise. These images or input signals need to be amplified to get a crisp image, but when amplifying these input signals, the noise in them gets amplified, making the image more distorted and harder to analyze.
Jerusalem, Israel-based Visionary.ai develops AI-based software algorithms that process image-based inputs to make them ready for analysis.
Read the full article here