Is Computer Vision also AI?
They've been a lot of controversy as to if computer vision is AI or not. They've claimed that computer vision is not Artificial Intelligence while others say it is. Some systems have been made and called AI so people say it?does not have enough grounds to be classified as artificial intelligence. Computer vision may be a sophisticated system but is it enough to be called AI? In this article, we will buttress this question and?clear all doubt.
Technology is a constantly growing field causing a swamp of new terminologies every year. Every successful technology that comes up tends to create new subfields and contexts sometimes making older technologies more complex to?clearly understand. The world keeps expanding with technological inventions and new ideas.
It is due to this pool of terminologies that brought mixed-ups in terms like automation and intelligence. This is when people see mere automation and call it intelligence. People see a bottling robot in a company exhibiting only automation and call it artificial intelligence. Computer vision on the other hand is like mimicking the human sense of sight, this is like automation without any necessary intelligence. Just like heat sensors in computers sensing heat levels. These systems just have the ability to receive data that is then processed. How this data is processed is what makes it AI or not AI.
What then is AI?
I do not want to define AI here using too much formality if not we will find it more difficult to rest our case. Understanding AI clearly will help us settle loads of our problems. Artificial intelligence is not just when the computer can perform on its own. It is not just when a system is highly sophisticated or cutting edge. What then is intelligence?
Artificial intelligence is seen when human intelligence is introduced at whatever level of tech sophistication. It can be seen when intelligence meets with automation or any other tech system, computer vision included. This implies mimicking the human way of reasoning no matter how small.
The core feature of AI is for a system to be able to respond in unseen situations. Responded even when they did not see it coming, neither them nor even the human programmer that programmed them. Just for a robot lifting its leg does not make it AI, but if it can use its legs to play football, from different new unforeseen angles and directions, maybe even learning new skills from previous games, then we can call it intelligent.
A repetitive task could just be done with automation. Automation does not settle for when the environment is loosened. New situations will be a failure. Where variables change within a more continuous domain.
For instance, a robot playing table tennis or football. The ball will keep coming from diverse angles and speeds requiring different responses from the robot. Trying to pre-program the whole possible angles and speed at which the ball could possibly come will not be a feasible approach. Intelligence could just use some learning approach to tackle this problem.
An intersection of intelligence and automation can be called AI. Now, how about computer vision? What makes a system AI is not if it collects data or how it collects the data, it is in how it processes it. Computer Vision only collects data through the camera, how this data is responded to or processed by the system is what makes it AI or not.
What is Computer Vision?
CV is a term that appears self-explanatory, yet it is many have gotten it wrong causing arguments and disagreements. The thing is so many people are too quick to use the term AI. Some have done this for marketing reasons, to please the buyer. Others have quickly concluded a system is AI by seeing its sophistication or how it performs a state-of-the-art task.
This hastiness is what has happened in my CV. The technology and field look very technical and attractive making it look good enough to be called AI. It is important to note that a CV can be AI or not depending on how it is used. CV is just a tool on its own. Using it may or may not be automation. This is the same as automation. A robot can be able to move on its own and still not be the intelligent same way a computer could see and still not be intelligent.
The problem is many people will define these terms without looking at them carefully. CV is the utilization of means which makes physical visualization possible for computers. Making use of means such as high-definition digital cameras. This feature could be very handy in generating or developing intelligent systems, it could be a tool for AI but not on its own.
领英推荐
A rough look at How CV Works
Summarily, let's see three things. The first is a camera. A camera is simply a simulation or automation of the human eyes. An artificial eye using artificial lenses. A camera is simply a device, but a step further will be a CV. CV is a field that takes this ability to CV to make systems that mostly also include intelligence. It makes AI very handy. CV uses digital cameras to capture data that is then processed. How this data is processed could be hard coding meeting computer functions and conditionals or it could be for making smart and intelligent systems.
CV could just be the processing of pixels as shapes or colors or patterns using simply mathematical arrays or pixels. This could just be graphics processing and setting some tight and refined conditionals. This is not AI yet. From a detailed look, CV could help in automation and still not be AI. From a lame man's view, it could just be summarized as AI since it performs actions like humans, but does it reason? If not, it is not AI yet.
Just sensing is not AI. CV could just be a sensor for sight. Even systems with heat sensors like the human skin are not AI yet until they can use the data sensed for intelligent decisions and reasoning in new circumstances like a human would do. Computer vision and other sensing abilities are often powerful tools for AI because they provide human ways of receiving data from the environment.
Conclusion
When should a system be called CV? A system is called CV if it principally focuses on Computer vision. If the main goal of developing the system is for CV, this includes even when the system has AI. We could have a CV facilitating automation and exhibiting intelligence. All these differences could be ignored. From an outward look, any sophistication containing automation or a CV should just be left as AI. But to be professional, and to practice the fields, it is very important to call a spade a spade if not implementation approaches will not be clear.
Key takeaways:
links:
Image by?Peace,love,happiness?from?Pixabay
Human Resource Specialist/ Recruitment Manager
2 年Quite educative?