Image Recognition is Computer Vision


Your eyes don’t just see, they understand. Actually, it is the brain which deciphers the images coming in through the eyes but the process is always includes recognizing what is seen. In the same way, a computer sees and understands through image recognition technology.

Image recognition is just another name for “computer vision”. It is the process through which a computer or website acquires, analyzes, and understands images and text. Once the image is recognized, the image recognition technology converts it into a numerical value for processing. It seems rather simple but in actuality is quite complicated.

Although computers are able to process information at a remarkable rate, its “vision” pales in comparison to the human eye. After all, the eye and brain functions have had millions of years to evolve. With that much time to refine the whole perceiving process, a human can recognize images almost intuitively. Even if images are 90 to 99% similar, the human eye picks it up and the brain automatically understands the slightest of differences. The result is the recognition of the letter “N”, for instance, over the letter “M”.

While the computer may not be able to see and understand at the level of the human mind, it does follow a similar pattern of processing. Image recognition technology is able to recognize what is on the screen through three processes: facial recognition, optical recognition, and pattern recognition.

Facial Recognition: This is the ability to distinguish one face from another. An example of this is when you upload and tag an image on Facebook and Facebook is able to distinguish it from all the other pics you have previously uploaded.

Optical Character Recognition (OCC): This technology is responsible for detecting and reading text. Have you ever deposited a check by simply taking a picture of it? If so, you have come across OCC technology. Your bank’s image recognition software is able to read the routing and checking numbers through the pic you send and therefore can deposit your funds into the correct account.

Pattern Recognition: This technology goes beyond the above two since it is not limited to any particular set of images or text. Pattern recognition can recognize any and all types of pixels and patterns ranging from the most generic to the most specific. If you are an avid Internet user, you know that you can now send a pic to Google and you will be supplied with a wide range of similar pics.

Both retailers and E-tailers can benefit from image recognition technology, specifically pattern recognition technology. One such company that provides visually recognized products through image recognition is Slyce.

Slyce uses image recognition technology to activate recognition for existing products as well as new products uploaded as photos. Their entire process entails three simple steps, which include a barcode, a photo, and a set of defining attributes. Slyce’s image recognition technology is suitable for both mobile and desktop devices.

To view a demo of image recognition: