Illustration by Ainsley Wagoner
For many decades, people dreamed of creating machines with the characteristics of human intelligence, those that can think and act like humans. One of the most fascinating ideas was to give computers the ability to “see” and interpret the world around them. The fiction of yesterday has become the fact of today.
Thanks to advancements in artificial intelligence and computational power, computer vision technology has taken a huge leap toward integration in our daily lives. The computer vision market is expected to reach $48.6 billion by 2022, making it an extremely promising UX technology.
In this article, we will review the concept of computer vision, discuss how this technology evolved, and share a few excellent examples where this technology can be applied in our lives.
What is computer vision?
Computer vision is the field of computer science that focuses on creating digital systems that can process, analyze, and make sense of visual data (images or videos) in the same way that humans do. The concept of computer vision is based on teaching computers to process an image at a pixel level and understand it. Technically, machines attempt to retrieve visual information, handle it, and interpret results through special software algorithms.
Here are a few common tasks that computer vision systems can be used for:
- Object classification. The system parses visual content and classifies the object on a photo/video to the defined category. For example, the system can find a dog among all objects in the image.
- Object identification. The system parses visual content and identifies a particular object on a photo/video. For example, the system can find a specific dog among the dogs in the image.
- Object tracking. The system processes video finds the object (or objects) that match search criteria and track its movement.
How does computer vision work?
Computer vision technology tends to mimic the way the human brain works. But how does our brain solve visual object recognition? One of the popular hypothesis states that our brains rely on patterns to decode individual objects. This concept is used to create computer vision systems.
Computer vision algorithms that we use today are based on pattern recognition. We train computers on a massive amount of visual data—computers process images, label objects on them, and find patterns in those objects. For example, if we send a million images of flowers, the computer will analyze them, identify patterns that are similar to all flowers and, at the end of this process, will create a model “flower.” As a result, the computer will be able to accurately detect whether a particular image is a flower every time we send them pictures.
Golan Levin, in his article Image Processing and Computer Vision, provides technical details about the process that machines follow in interpreting images. In short, machines interpret images as a series of pixels, each with their own set of color values. For example, below is a picture of Abraham Lincoln. Each pixel’s brightness in this image is represented by a single 8-bit number, ranging from 0 (black) to 255 (white). These numbers are what software sees when you input an image. This data is provided as an input to the computer vision algorithm that will be responsible for further analysis and decision making.
The evolution of computer vision
Computer vision is not a new technology; the first experiments with computer vision started in the 1950s, and back then, it was used to interpret typewritten and handwritten text. At that time, computer vision analysis procedures were relatively simple but required a lot of work from human operators who had to provide data samples for analysis manually. As you probably guess, it was hard to provide a lot of data when doing it manually. Plus, the computational power wasn’t good enough, so the error margin for this analysis was pretty high.
Today, we do not have any shortage of computer power. Cloud computing, paired with robust algorithms, can help us solve even the most complex problems. But not just the new hardware paired with sophisticated algorithms (we will review them in the next section) is driving computer vision technology forward; the impressive amount of publicly available visual data that we generate every day is responsible for the recent process of this technology. According to Forbes, users share online more than three billion images daily, and this data is used to train the computer vision systems.
Deep learning revolution
To understand the recent process of computer vision technology, we need to dive into algorithms this technique relies on. Modern computer vision relies on deep learning, a specific subset of machine learning, which uses algorithms to glean insights from data. Machine learning, on the other hand, relies on artificial intelligence, which acts as a foundation for both technologies (check AI design best practices to learn more about design for AI).
Deep learning represents a more effective way to do computer vision—it uses a specific algorithm called a neural network. The neural networks are used to extract patterns from provided data samples. The algorithms are inspired by the human understanding of how brains function, in particular, the interconnections between the neurons in the cerebral cortex.
At the core level of a neural network is the perceptron, the mathematical representation of a biological neuron. Similar to biologic neurons in the cerebral cortex, it’s possible to have several layers of interconnected perceptrons. Input values (raw data) get passed through the network created by perceptrons and end up in the output layer, which is a prediction, or a highly educated guess about a certain object. For example, at the end of the analysis, the machine can classify an object with X% confidence.
Where we can apply computer vision technology
Some people think that computer vision is something from the distant future of design. Not true. Computer vision is already integrated into many areas of our life. Below are just a few notable examples of how we use this technology today:
Computer vision systems already help us organize our content. Apple Photos is an excellent example. The app has access to our photo collections, and it automatically adds tags to photos and allows us to browse a more structured collection of photographs. What makes Apple Photos great is that the app creates a curated view of your best moments for you.
Facial recognition technology is used to match photos of people’s faces to their identities. This technology is integrated into major products that we use every day. For example, Facebook is using computer vision to identify people in photos.
Facial recognition is a crucial technology for biometric authentication. Many mobile devices available on the market today allow users to unlock devices by showing their faces. A front face camera is used for facial recognition; mobile devices process this image and, based on analysis, can tell whether the person who is holding a device is authorized on this device. The beauty of this technology is that it works really fast.
Computer vision is a core element of augmented reality apps. This technology helps AR apps to detect physical objects (both surfaces and individual objects within a given physical space) in real-time and use this information to place virtual objects within the physical environment.
Computer vision enables cars to make sense of their surroundings. A smart vehicle has a few cameras that capture videos from different angles and send videos as an input signal to the computer vision software. The system processes the video in real-time and detects objects like road marking, objects near the car (such as pedestrians or other cars), traffic lights, etc. One of the most notable examples of applications of this technology is autopilot in Tesla cars.
Image information is a key element for diagnosis in medicine because it accounts for 90 percent of all medical data. Many diagnoses in health are based on image processing—X-rays, MRI, and mammography, just to name a few. And image segmentation proved its effectiveness during medical scans analysis. For example, computer vision algorithms can detect diabetic retinopathy, the fastest-growing cause of blindness. Computer vision can process pictures of the back of the eye (see below) and rate them for disease presence and severity.
Cancer detection is another notable example. Accuracy in diagnosing different forms of cancer is vital. According to Google, computer vision tools assist in detecting cancer metastasis with much higher precision than human doctors. Below you can see a closeup of a lymph node biopsy. The tissue contains a breast cancer metastasis as well as areas that look similar to the tumor but are benign. The computer vision algorithm successfully identifies the tumor region (bright green) and is not confused by the normal areas that look like tumors.
Many agricultural organizations employ computer vision to monitor the harvest and solve the common agricultural problems such as weeds emergence or nutrient deficiency. Computer vision systems process images from satellites, drones, or planes, and attempt to detect the problems in the early phase, which helps to avoid unnecessary financial losses.
Computer vision is a popular topic in articles about new technology. A different approach to using data is what makes this technology different. Tremendous amounts of data that we create daily, which some people think as a curse of our generation, are actually used for our benefit—the data can teach computers to see and understand objects. This technology also demonstrates an important step that our civilization makes toward creating artificial intelligence that will be as sophisticated as humans.