Artificial intelligence has become one of photography’s best coworkers. Every time we take a photo with a mobile we are using that intelligence that processes the data and allows the resulting photography to be of better quality. Sony has wanted to take this symbiosis to a higher level.
The technology company specialized in photography has created two new image sensors with artificial intelligence integrated as standard. In this way, they avoid depending on the internet connection to be able to talk to intelligent systems when taking a photo.
The problem that Sony has wanted to solve with the new IMX500 and IMX501 implies that the artificial intelligence programs that assist the cameras are not found within the devices themselves, but thousands of km away on large servers. Only through the Internet connection can they communicate with each other, which implies a series of drawbacks.
Having to send the data collected by the camera sensor to the server where the AI is located involves a small latency, very small, but that could be avoided if we integrated the AI into the device. It also means that our data has to travel outside the device and be stored in the cloud, a system that could present security breaches and not comply with data protection laws such as the GDPR in some cases.
Third, the fact of having an intelligent system in the sensor itself eliminates the need to send this data to a DSP (Digital Signal Processor) housed in the device. One less element that we have to install, giving more free space to the rest of the camera or mobile components.
All these problems would be solved with the new structure of the Sony IMX500 and IMX501, which are made up of two layers. At the bottom is the artificial intelligence and memory engine, while at the top is the image sensor that collects light and all the data from the scene so that the AI can process them. According to Sony, the process takes just 33ms to complete.
Not only can mobile cameras benefit from these new sensors, but many cameras also require intelligent processing of the images they collect. From the security cameras, we use at home to the ones we have in cars to make them more autonomous. In addition to improving the quality of images, artificial intelligence can analyze and recognize objects captured by the camera.
Sony emphasizes in its announcement that any device that requires image processing and intelligent recognition can take advantage of these sensors to boost their functions. Even the smart speakers we have at home could use it to recognize the face of the person who is giving them the orders or examine an object that we want to search the internet to add to the shopping list.