How computer vision and AI can impact industrial production

In this article we will see about industrial production with computer vision and artificial intelligence, what hardware and software are needed, its applications within the industrial production process, from the beginning of production to its disassembly, and in the end overall analysis and what are the challenges in implementation and use.

The main components for industrial production with Computer vision and AI

Before proceeding with a description of how computer vision impacts industrial production, we need to make a quick description of the basic hardware and software components needed to perform this process: camera, light, software and hardware

Camera and light

A key aspect of this process is the choice of camera, and about this I invite you to read How To Measure The Size Of Objects With A Very High Accuracy . The camera captures the image of the object to be monitored, each frame is then sent to the processing device and through the software gives us the required information.

Another aspect to be strongly considered is light. A controlled environment with uniform light prevents shadows or color changes on the object to be checked

Computer vision system for industrial production

Software and hardware

Once captured, the image is sent to a device for example Jetson Nano, Computer, or Server In this way, computer vision software with artificial intelligence can derive the information, which may be about inspection, classification, tracking, or other.


If the software for industrial production consider that action is needed based on the result of frame processing, it instructs a hardware device to proceed to iminate the object from the production cycle if defective, or seperate it based on category, etc.

robot arm computer vision

Manufacturing processes with computer vision

Now that we have seen what a manufacturing inspection module with computer vision consists of, we can analyze the entire production cycle from product design to disassembly.

1. Design

Step to create a product or digital copy of the product. Creating a 3D representation of the object helps in design verify or test the product with much-reduced costs compared to making them in a real environment. With computer vision, one can capture the image digitally of the object by having a precise map of all its points, and this will also facilitate its processing.

2. Scheduling and planning

After anlizing the object we can analyze and plan its processing to arrive at the finished result. Let’s take the example of this log, we can know exactly where the knots in the wood, various defects, damaged parts are, and during the processing phase we can indicate with precision to the robot which points ( giving the precise coordinates) should be discarded.

wood defect

3. Production process

During production, it may be necessary to track the object as it moves through a conveyor belt. For example if we were facing the production cycle of glowing materials it may be necessary to track objects and in these cases the only possible way is with computer vision, we see a description of it in this article Object Tracking From Scratch – OpenCV And Python . If you want to know more about metal production I recommend “METAL FORMING PROCESSES: INDUSTRIES & APPLICATIONS

Computer vision tracking

4. Inspection

This is the most interesting aspect for industrial production because product inspection is the first thing people think of when they talk about computer vision. Inspection is about quality control, classification and defect detection.

computer vision opencv industrial production inspection

5. Assembly

Computer vision is also used in assembly such as providing the coordinates of the spot to be welded or checking for missing electronic components

circuit assembly

6. Trasnsportation

Another key aspect for industrial production is logistics. One can use self-driving vehicles that detect the route and autonomously manage movements or track vehicles moving within a warehouse for safety purposes.

7. Disassembly

Computervision is the same as used for robot guidance to help with disassembly of objects. For example, it can help the operator by highlighting the parts to be removed or autonomously proceed with the extraction of screws or bolts.

Is it an advantage for companies to implement computervision?

Regarding reliability, there is no doubt that it is definitely valuable to implement. Whereas as for whether it is for all companies, this is not obvious. The implementation of computerviosion has a high cost, and it is always good to calculate the return on investment before proceeding.


Learn to build Computer Vision Software easily and efficiently.

This is a FREE Workshop where I'm going to break down the 4 steps that are necessary to build software to detect and track any object.

Sign UP for FREE