Not long ago, engineers designing quality control systems had to choose from several inspection options, none of which were entirely satisfactory. These options included expensive single-purpose vision systems, multi-array low-functionality photoelectric sensors, and human eye inspection, which was susceptible to fatigue and lack of concentration. However, today's state-of-the-art vision sensors are combining the best performance of traditional methods with unprecedented speed, accuracy, size, and cost advantages. This article compares the capabilities of various sensing options and demonstrates why vision sensing is the fastest-growing sector in the industrial sensor market. The Basic Principles of Vision Sensing A photoelectric sensor contains a light-sensing element, while a vision sensor has thousands of pixels that capture light from an entire image. The sharpness and detail of an image are typically measured by resolution, expressed in pixels. Some vision sensors offered by Banner Engineering are capable of capturing 1.3 million pixels. Therefore, the sensor can "see" a very detailed image of a target, whether it is meters or centimeters away. After capturing the image, the vision sensor compares it to a reference image stored in memory for analysis. For example, if a vision sensor is programmed to identify a machine part with eight bolts correctly inserted, the sensor knows it should reject a part with only seven bolts, or a part with misaligned bolts. Furthermore, the vision sensor can make this judgment regardless of the machine part's location within the field of view or whether it has rotated within a 360-degree range. Advantages of Vision Sensors Among available inspection alternatives—vision systems, photoelectric sensors, manual inspection, and vision sensors—vision sensors are generally the best choice due to their accuracy, ease of use, rich functionality, and reasonable cost. As competition intensifies and profit margins shrink across industries, manufacturers cannot afford high scrap rates caused by defective products. Therefore, to detect problems before incurring high costs, manufacturers are integrating inspection into the entire manufacturing process. ◆ Comparison of Vision Systems and Vision Sensors Complex vision systems are a mature technology capable of performing detailed automated inspections. However, their complexity and high cost, typically ranging from $5,000 to over $50,000, hinder their application in many industries. These complex vision systems require one or more cameras, custom software, and a computer. These systems often require external vision consultants for design, integration, and installation. Furthermore, due to their specialized nature, they cannot be easily repurposed. These complex systems typically require ongoing professional support. While the demand for complex vision systems remains, the introduction of cheaper and easier-to-use vision sensors has provided more cost-effective solutions for some industrial applications. Moreover, due to their smaller size and ease of use, manufacturers are adopting vision solutions more frequently in inspection and verification applications. Vision sensors have played a crucial role in improving quality and productivity in factory automation. ◆ Comparison of Photoelectric Sensors and Vision Sensors Compared to photoelectric sensors, vision sensors offer machine designers greater flexibility. Applications that previously required multiple photoelectric sensors can now be inspected using a single vision sensor. Vision sensors can inspect much larger areas and achieve greater flexibility in target position and orientation. This has made vision sensors widely popular in applications that were previously only possible with photoelectric sensors. Traditionally, these applications also required expensive accessories and precise motion control to ensure that the target object always appears in the same position and orientation. Furthermore, since the cost of a basic vision sensor is equivalent to only a few photoelectric sensors with more expensive accessories, price is no longer an issue. Vision sensors offer unparalleled flexibility for application switching. For example, switching a production process (from single-serving yogurt to ice cream tubs) can take only seconds and can be done remotely. Additional inspection conditions can be easily added to this application. ◆ Comparison of Manual Inspection and Vision Sensors Regardless of advancements in factory automation, many inspections are still performed by the naked eye. However, in most applications, vision sensors offer numerous advantages that manual inspection processes cannot match. Vision sensors can operate at much higher speeds and perform repetitive, multiple, and consistent inspections at much lower costs. Expanding Application Range The low cost and ease of use of vision sensors have attracted machine designers and process engineers to integrate them into a variety of applications that previously relied on manual labor, multiple photoelectric sensors, or no inspection at all. Industrial applications of vision sensors include inspection, metrology, measurement, orientation, defect detection, and sorting. Here are just a few examples: ◆ In an automotive assembly plant, inspecting whether glue beads applied by a robot to the door frame are continuous and of the correct width. ◆ In bottling plants, verifying correct bottle cap seals, filling levels, and preventing foreign objects from falling into the bottle before sealing. ◆ On packaging production lines, ensuring correct packaging labels are affixed in the correct positions. ◆ On pharmaceutical packaging production lines, inspecting blister packs of aspirin tablets for broken or missing tablets. ◆ In metal stamping companies, inspecting stamped parts at a rate exceeding 150 tablets per minute, more than 13 times faster than manual inspection. Conclusion: This is an exciting time for vision sensors. Technology that once required extensive expertise is now economical and easy to use. Future product development using this technology will continue this trend. The current challenge is to make various industries aware of the potential of vision sensors in all manufacturing sectors, including quality control, measurement, and inspection processes. Machine Vision Improves Quality Control Efficiency For factory automation and process automation, machine vision is the foundation for achieving true automation and an important means of quality control. Below are some of Omron's views on the domestic machine vision market. Machine vision, as the name suggests, uses automated methods to achieve functions similar to human vision (eyes + visual nerve center + visual nerve cells). For factory automation and process automation, machine vision is the foundation for true automation and a crucial means of quality control. In its simplest terms, a machine vision system consists of a lens, camera, and controller. It replaces manual labor by measuring and judging various parameters such as size, defects, type, matching, and text based on images of objects in a specific environment. This helps customers improve production efficiency and reduce the impact of uncertainties inherent in manual labor on product quality. The current situation in China regarding the entire machine vision industry chain is somewhat unique. On the supplier side, since the vast majority are foreign manufacturers, there is a challenge in understanding the Chinese market and culture. Everyone has witnessed China's rapid economic development, but the driving factors may differ from those abroad, which is something they cannot fully comprehend in the short term. It is believed that the current sales situation in China is still far from the expectations of most suppliers. Even for companies like OMRON, which has been in the Chinese market for over two decades and has experienced rapid growth in machine vision in recent years, its current market sales are still far behind those in Japan. As for users, the biggest problem lies in their lack of understanding of machine vision. This is partly due to the unique circumstances in China, namely the comparison between domestic labor costs and the initial investment in machine vision. This could stem from a lack of understanding of new products or problem-solving methods, or a unclear comparison between this new method and traditional manual methods. Suppliers need to provide more support. Similarly, system integrators, the bridge between users and suppliers, also lack understanding of machine vision, unaware that besides mechanics and ordinary sensors, there's another method: making judgments through images/visual representations. There's also a significant shortage of domestic system integrators, and the proportion of domestically manufactured equipment is too low—a problem that cannot be solved in the short term. Given the relatively low labor costs in China, the initial investment in machine vision is undoubtedly a crucial consideration for anyone considering purchasing it, placing higher demands on machine vision systems in terms of both price and performance. Based on these understandings, industry experts suggest that early market cultivation, customer education, and specifically, the cultivation of system integrators are extremely important during the growth stage of this new product. Therefore, we have invested considerable time in training on the fundamentals of machine vision, hoping to first ensure that more people understand machine vision and its basic principles. The future of machine vision is very optimistic and full of confidence. We believe that with continuous improvement, as domestic automation levels increase, societal demands for product quality rise, and manufacturing industries demand higher production efficiency, machine vision—this new product and concept—will gain widespread acceptance. Just like the stages that today's leading automation nations went through, machine vision will achieve even greater success in a larger market. Therefore, OMRON believes the future of machine vision is bright, though the road is long. As for the specific timeframe, it might take 5 years, or it might only take 3 years, or even less, because once domestic development begins, the speed will be astonishing. Postscript: A prominent problem in China's food and packaging machinery industry is the long new product development cycle, with much imitation and few new ideas. This is not only related to the knowledge level of designers but also to the development of related industries. The internationally advanced approach is to store various machine elements in a database in a computer, then digitize the drawings and input them into the computer, which automatically synthesizes a 3D model. Next, past production line failures are input into the computer, which can then demonstrate the actual working conditions and allow users to make modifications as needed. Computer synthesis is fast, and modifications are quick and convenient. If certain user requirements are not met, the computer will also inform you to avoid future disputes. This computer-generated simulation technology greatly shortens the design cycle of packaging machinery (production lines), and significantly improves customer satisfaction. Of course, the application of this simulation technology also reminds us that the manufacturing of packaging machinery should be modularized according to the function of the components, with each component having a unified interface and modules that can be freely combined. This can transform the small-batch production of single pieces of packaging machinery into mass production of various modules, thereby reducing production costs. Even if it is not a production line but a single machine, modular combination can shorten the design and manufacturing cycle and reduce production costs. Furthermore, modularization can standardize non-standard parts, making troubleshooting easier and faster for users, thus reducing their operating costs.