What Computer Vision Can Do That Human Vision Can’t

By Matt Thurston

Computers can’t “see” in the way that we humans can. More importantly, computers don’t have the consciousness and emotions required to form conceptions and contexts about what they see in the way that we humans can. Nevertheless, “computer vision” can – and does – do some things better, not to mention more efficiently, than we humans do.

What is Computer Vision?

According to Wikipedia, computer vision is “an interdisciplinary scientific field that deals with how computers can gain high-level understanding from digital images or videos. From the perspective of engineering, it seeks to understand and automate tasks that the human visual system can do.”

Originally a somewhat pie-in-the-sky idea proposed by computer engineers and programmers back in the 1960s, computer vision has come a long way since then. For instance, using a combination of advanced camera technology, artificial intelligence (AI) concepts, and “deep learning” methodologies, computers can now outperform human beings when it comes to such things as classifying an image – is it a horse or a zebra? – and analyzing a medical image, such as those produced by an MRI or a CT scan, to detect anomalies that indicate the existence of an illness or injury and its extent.

Seeing by Different Rules

The most important thing to know about computer vision is that it works completely differently than human vision. It relies on the computer’s neural network, called a “feed-forward” system, that allows the computer to perform the following series of steps:

• Receive an image
• Detect the image’s pixels
• Detect its edges and contours
• Analyze the entire object
• Make an “educated” guess about what the image is

Scientists and medical researchers know that human vision depends on each person’s visual cortex, the area of the brain in which he or she creates an accurate and very vivid representation of objects in the world around him or her based on the relatively scant information that comes in through the retinas in his or her eyes.

But exactly what’s happening and how it’s happening is still a mystery. For example, what causes us to do a double-take? And how does the information we receive from our second glance tell us what’s happening? For another, why is the conclusion drawn by a 10-year-old so different than that drawn by a 40-year-old? The answer to that question is that our vision matures over the years. Some experts say that we can’t fully zero in on what we’re looking for, suppressing the surrounding clutter, until we reach age 17. Other experts say that our ability to fully perceive other human faces doesn’t mature until at least age 20. Computer vision lacks this maturation ability.

Important Computer Vision Applications

Despite the limitations of computer vision, however, it already serves some very important real-world functions in addition to health care imaging. Some of these applications include:

• Manufacturing line assembly and inspection
• Traffic control systems
• 3D printing
• Pharmaceutical package inspection
• Forensic investigations
• Automotive safety features
• Military and defense pattern recognition and image processing
• Search and rescue operations
• Fingerprint scanning systems
• Home and business identification and access control

You’re likely already using computer vision on a personal level. How? If you use Google Photos, computer vision is what allows your photos to be classified in terms of their content. In other words, computer vision relieves you of the necessity of adding tags and descriptions to each photo. It’s also what allows you to quickly search through your photos or videos by simply typing in the type of content you’re looking for.

Your smartphone and other mobile devices likewise use computer vision for such things as:

• Online shopping
• Barcode scanning
• Speech recognition and speech-to-text capabilities
• Optical character recognition

These capabilities are especially helpful if you’re vision impaired.

Bottom line, computer vision is here to stay and likely will become even more amazing in years to come.

Matt Thurston: I’ve been working in the tech industry for more than 15 years. I have always loved sharing the insights that I have discovered throughout my career with other tech and business enthusiasts. I have a passion for writing and enjoy sharing my findings with others. My writing is informational in nature and provides readers with innovative content.