machine-vision

Democratization of machine vision and image processing

Machine vision and image processing are at a major turning point in the FA / automation industry. In addition to technological evolution such as 3D and AI utilization, it has become easy and easy to use, and new market development is expected.

Eyes of machines and appliances

As you may know, machine vision is, as the name suggests, “machine eyes”, which means sensing that processes and uses images taken by so-called industrial cameras. There are four main uses, positioning, identification, measurement, and inspection, and the acronym is called “GIGI” (Guidance = positioning, Identification = identification, Gaging = measurement, Inspection = inspection).

Since machine vision is the “eye”, it is possible to collect a lot of easy-to-understand and useful information such as images, which is very useful for adding value to the operation of machines and equipment. The amount of information input to the brain is completely different between walking with eyes closed and walking with eyes open, and just as walking speed, accuracy, and safety are significantly different, machine vision is applied to machines and devices. By utilizing it, it is possible to greatly improve the performance. In addition, the market is expanding year by year as it enables us to perform tasks that were previously impossible.

According to a survey by Fuji Keizai, the global market for image processing systems will be 1,283.6 billion yen in 2019, which will grow to 1,502.4 billion yen in 2022. For several years, it has become a global trend to introduce machine vision and image processing systems for labor saving and automation of visual inspection. It is easy to think that it is already widely used, but it is not surprising, and it is certain that it will continue to grow due to the expansion of the robot market and the rise in demand for automation due to the corona disaster.

The challenges to popularization are technical difficulty and high cost

On the other hand, there are also challenges. Machine vision and image processing are technically difficult and costly to install and use.
In order to create a system that can be recognized with high accuracy and can withstand work, it is not just a matter of buying and installing an industrial camera. It is necessary to combine hardware such as cameras, lenses, and lighting, create image processing software, optimally install and install, and make fine adjustments to these.

The accuracy varies greatly depending on the focus and the way the light is applied, and when it comes to building software, it is extremely difficult for an amateur to do it. Therefore, when introducing an image processing system, it is common to ask a specialist who is familiar with vision technology such as a manufacturer who is strong in image processing or a system integrator, which tends to increase the cost, which is a problem for users who want to introduce it. It has become a point. In particular, the difficulty and cost of technology tends to increase as image processing and technology using 3D and AI become more sophisticated.

To image processing that anyone can use. Democratization of machine vision

“Image processing is difficult and expensive” This was the conventional wisdom, but recently the trend has changed. Machine vision technology trends are now polarized. One is the so-called “advancement” that enhances recognition accuracy and speed by refining AI and 3D technologies that are an extension of conventional technology. The other is “generalization” that makes it easier for anyone to use in order to spread machine vision more widely. To put it coolly, it may be called “democratization of machine vision.” In particular, the latter is gaining momentum, and many manufacturers have introduced machine vision that is easy to introduce, which has been a tailwind for the automation of visual inspections and the spread and expansion of machine vision.

The image processing market is expanding rapidly in the future. Now is your chance!

Nowadays, it is common to take pictures and images with smartphones and enjoy them, but it is only in the last two or thirty years that cameras and images have become a part of our daily lives. Taking pictures with a camera is an area of ​​professionals and camera enthusiasts, and the timing of shooting for the general public was only during extraordinary times such as travel and events. With the evolution of technology from silver salt cameras to compact cameras, disposable cameras, digital cameras, mobile phone camera functions, and smartphones in a general-purpose direction, it has become so popular that it has become so popular and has various ways of enjoying and new photography. Culture was cultivated.

Machine vision and image processing systems are the same, and until now they were limited to the area of ​​specialists due to technical difficulties, but now they can do it easily and have expanded to the area of ​​users. .. In a little more time, it will become more versatile and various applications will come out.

For people who want to automate visual inspections, use image processing to streamline their work, or start a new business with machine vision, it may be an opportunity now that generalization has begun.