In modern factories, where production lines operate at exceptionally high speeds, even the smallest defect can become a major problem. A tiny crack, an electronic component that is not precisely soldered, an unreadable or poorly marked barcode, a faulty raw material, dirt, or a small deviation in the structure of a medical product – all of these can lead to product rejection, economic losses, and even safety risks.
As industrial production becomes more complex and faster, the need for systems that can "see" and react quickly and consistently grows.
This is precisely where 1Vision, an Israeli company founded in 2008 by three partners, enters the picture, quite literally.
The company specializes in developing automated quality control systems based on image processing, electro-optics, and AI. The company's systems are installed on production lines across a wide range of industries: defense, medical, electronics, agritech, and more, enabling the identification of deviations and defects in real time during production.
Among the three founders of the company, we met CEO Gilad Horvat, who holds an MBA from Tel Aviv University and was formerly a physics engineer, having been involved in development during his military service and later in industry. We also met Andre Elazary, VP R&D, who holds an M.Sc. in Management of Technology from NYU and B.Sc. in Electrical Engineering from the Technion, and came from a similar professional background. The third partner is the CTO Erez Shapira.
The three partners knew each other before founding the company, having worked in the field of quality control and industrial cameras, which was then in its early stages. "We wanted to develop inspection systems at a time when there was almost no market for it," says Horvat. "From the experience we gained, we understood that it's possible not only to identify defects but also to completely change the way factories operate."
Elazary adds that from the outset, it was clear that the real challenge was not just choosing a camera or developing software but deeply understanding the production environment itself and the changing needs of each industry.
The Precision Behind the Image
Since its establishment, 1Vision has operated on two main fronts: developing customer-tailored systems, alongside commercial activity that includes marketing and distributing technologies and products from leading international manufacturers. This combination, according to the two, is one of the company's greatest advantages.
"We work with the most advanced suppliers in the world and receive information from them about technologies that will only be released in another year or two," says Horvat.
"This exposure allows us to integrate the future into the developments we are making today.
" According to him, the company's knowledge center relies on two main areas: image processing and careful selection and integration of imaging hardware. "It's not enough to know how to write an algorithm," he explains.
"You also need to know how to choose the right camera, the appropriate lens, and the precise lighting for each object. Sometimes a small change in lighting or wavelength can make defects significantly more prominent in the image. Many years of experience are a crucial parameter."
How does it work?
Elazary: "1Vision's systems are based on industrial cameras, optics, illumination and image-processing algorithms that capture images of products as they move along the production line, analyze the acquired image data in real time, and decide whether the product is conforming or defective. If a deviation is detected, the product is automatically ejected from the line, or the system sends an immediate alert. This ability to perform all checks during the production process, without affecting the production throughput, is one of our main advantages. Most customers need inline solutions, so products are checked in motion, immediately filtered if there's a problem, and all information is saved so that the process itself can be learned from and improved."
In certain cases, and as needed, Horvat qualifies, "If the same defect appears repeatedly in the same area, some customers request to activate a red light or a trigger to stop the line, in order to correct the cause of the rejection, thereby saving significant waste of raw material and time."
The Connection Between AI and the Production Line
At 1Vision, they explain that the main challenge is not just developing advanced technology, but precisely adapting it to the specific domain, needs, and unique requirements of each customer and industry. Each field has completely different characteristics – a system designed for the medical field is very different from a system for the plastics or electronics industry.
"Every industry has its own standards, sensitivities, and challenges, so there is no one-size-fits-all solution," Horvat explains. According to him, it is precisely the ability to combine different technologies and develop customized solutions that allow for high levels of accuracy, without making the systems too complex or expensive.
One of the characteristics is the integration of AI models with traditional image-processing algorithms. This approach is based on the understanding that it is not always correct or efficient to apply artificial intelligence at every stage of the process. Some checks are performed using traditional methods that are sometimes faster or more accurate, while only areas or defects that cannot be precisely characterized by classical approaches are sent for analysis using AI modules. According to Horvat, this combination allows for maintaining a high work rate, while avoiding unnecessary waste of computational resources.
Horvat refers to the advantages of developing independent models, alongside marketing AI software from international companies: "The software we market has an advantage, based on thousands of hours of work invested in developing the models. This allows them to achieve a high level of accuracy by learning from a small number of images. In addition, they offer flexibility for changes when testing new products on the production line," he explains. "On the other hand, some customers want to embed an AI model for image processing directly into their product, especially when it is a product intended for large-scale production and sale. In such cases, 1Vision focuses on developing the AI model itself, rather than supplying a complete off-the-shelf product."
Behind 1Vision's developments is also a broader vision for the future of the industry. Thus, artificial intelligence is not just a technological addition, but part of a deeper process in which factories are gradually becoming more autonomous, precise, and faster. According to Horvat, the challenge is to connect several technologies: image processing, machine learning, databases, hardware, and software, in a way that truly works in the real world, at real speeds, and under real production conditions, while generating insights for management.
Elazary emphasizes that another advantage of AI is its ability to understand natural variation in products. "An electronic component can be slightly larger, smaller, have a different texture, or slightly crooked lines, and still be functional," he describes the importance of integrating advanced tools, which allows for understanding the permissible range and intelligently distinguishing between functional and defective. This is not just according to a rigid template, but thanks to process automation, similar to how a skilled human operator learns over time.
Although artificial intelligence has become a significant part of the solutions developed by 1Vision, Horvat and Elazary emphasize that in the industrial sector, reliability is a critical condition. According to them, in a medical factory, for example, it is unacceptable for a system to pass defective products or reject a high percentage of functional products. Therefore, exceptionally high levels of accuracy and consistency are required. The complexity in the medical field is not limited to defect identification but also includes the requirement for full traceability of every product. "If a medical implant is produced, it must be numbered and have full documentation that it passed all checks," Elazary explains. "This way, if a fault occurs in the future, the manufacturer must be able to prove that the product left the production line in good condition."
In this context, the company has also developed systems for inspecting syringes for biological drugs, products where any fault, even the smallest, can be critical. "There are very sensitive materials there," Horvat reveals, emphasizing that a small leakage is unacceptable, and therefore highly reliable and accurate systems are essential.
In other cases, the main challenge is precisely the speed of work, such as reading serial numbers or barcodes embedded in products. "In an automated system, you have one chance to capture images ofthe product," he explains. "If you don't succeed, you might get a false rejection. These are precisely the complex projects that come to us thanks to the knowledge and experience we have accumulated."
Future Production Lines
One of the company's first projects was for an electronics manufacturer that supplied equipment to Boeing. At that time, each component was manually inspected by four different employees with a magnifying glass. "Only if all four inspectors confirmed that the component was functional did it pass," Horvat recalls. "If one component was rejected, the entire shipment was rejected. That's where we came in with our first system."
Today, 1Vision's systems operate in a wide range of fields. In one project, for example, a system was developed for high-speed counting of microscopic insects for the AgriTech sector. In another project, in New Zealand, a system that was installed uses a thermal camera to check whether baby food containers were properly sealed.
The company's activity also crosses borders, with installations in Australia, New Zealand, the United States, India, Singapore, and Mexico. A significant part of the support is provided remotely. "Even during the war, there were cases where people could not reach the factories," Horvat says. "Even then, we were able to remotely control the systems, repair, change parameters, and update software, both in Israel and overseas."
The future, for them, is clear: more automation, more AI, and more smart systems that will be accessible to smaller factories as well. "In the coming years, we will see many more automated production lines," says Elazary. "This need is constantly growing, both due to labor shortages and increasing demands for quality and accuracy."
The company's next goal is to develop standardized or configurable product platform that will allow for wider distribution in international markets. "We aim to leverage the extensive knowledge and experience we have accumulated, along with the operational infrastructure that enables growth, to create replicable solutions with minimal adaptations for each customer," Horvat concludes. "The goal is to continue to be at the technological forefront and develop solutions that will make smart industry accessible on a global scale."
It is important to emphasize that the primary driver for automation in many industries today is not workforce reduction, but the growing shortage of personnel for repetitive manufacturing and inspection tasks. Automated vision systems provide consistent 24/7 operation, eliminate subjective inspection variability, and generate production data that can be used for continuous process improvement.
For manufacturers operating in competitive global markets, automation has become a strategic necessity. By combining machine vision, AI, robotics, and process analytics, factories can improve yield, reduce waste, increase throughput, and maintain high quality standards regardless of labor availability
For more info: 1vision.co.il
Partnered with 1Vision
