"Roxy AI," which realizes the automation of visual inspections, has been upgraded to version V1.13! An AI that has learned from inappropriate data is useless. Roxy AI maximizes AI performance by cleaning the data!

Garbage In, Garbage Out. An AI that has learned from incorrect data is useless.

There is an explanatory video, so please take a look. Garbage In, Garbage Out. An AI that has learned from incorrect data is useless. In fact, AI developers say they spend 80% of their time on data preparation. Can using the latest research results and the strongest algorithms solve the problem? No. Underestimating data will render even the strongest AI a wasted treasure. If you allow difficult-to-find defects mixed in with normal images to be learned, the result will be an AI that does not detect features similar to the mixed-in defects, and that is not the AI you expected. What a human can intuitively understand, AI cannot adapt to. So, should we thoroughly check all normal images to ensure that defects are definitely not present? That would be ideal, but it requires an enormous amount of time and attention. Roxy AI considers that inappropriate data is something that will mix in. Inappropriate data that has mixed in can be removed and cleaned up. With Roxy AI, data can be cleaned throughout the entire AI learning process. We clean the data and create conditions that make it easier for the AI to learn.

• Visual Inspection Equipment

The phrase "high precision," which is used casually, has a solid reason behind Roxy AI's high precision.

There is an explanatory video, so please take a look. The phrase "high precision" is often used casually. Ask various companies: What is the reason for your high precision? Roxy AI achieves high precision with three key commitments. Without any one of them, precision cannot be achieved. ■ A simple AI focused on necessary functions It can only detect defects and identify defect types. It cannot fill in defective parts or read text. Even so, it can meet the needs of many workplaces. Because the tasks are simple, it not only makes high precision easier to achieve, but also offers many advantages such as easy annotation and fast learning. ■ Learning the characteristics of both normal and defective products together No need for defective samples. This is indeed an attractive phrase. However, many people may feel there are limits to learning from normal products. By properly learning the boundary between normal and defective products, we aim for high precision. ■ A thorough process to eliminate inappropriate data An AI trained on incorrect data is useless. By organizing the training data, we create conditions that make it easier for the AI to learn. There are many features available for this purpose.

• Image Processing Software

The risk of foreign substances contaminating various food products. Using Roxy AI can help reduce such risks.

There is an explanatory video, so please take a look. In June 2022, shocking news emerged about the contamination of dried baby fish with pufferfish larvae. Fortunately, it seems that no serious harm was done, but it was a case that could have led to health issues with just one misstep. It may seem easy to distinguish between small fish and pufferfish. However, detecting foreign objects like pufferfish among a large quantity of small fish without missing any is a challenging task that requires constant vigilance. Similar incidents occurred in September 2014 and July 2019. It is believed that measures such as strengthening inspection processes have been taken each time, but relying solely on human labor cannot completely eliminate mistakes. The inspection processes for food products, which have had to rely on human labor until now, can be automated using Roxy AI. Automating inspections not only reduces the risk of foreign object contamination but also addresses many issues such as alleviating labor shortages, stabilizing quality, and improving the crowded environment of inspection areas. The video introduces examples of detecting various foreign objects among dried sardines. Please take a look.

• Food Testing/Analysis/Measuring Equipment

Roxy AI, which can be mastered even by non-engineers. See how it differs from other appearance inspection AIs.

Unless it is an extremely simple test, a perfect AI cannot be created in just one learning session, so it is very important to improve accuracy. Many AI inspection products create AI using one of the following methods: - Utilizing AI know-how to repeatedly tune parameters to improve accuracy - A simple one-click process for anyone. If the accuracy is poor, data is endlessly collected Just as studying from incorrect or ambiguous textbooks does not improve grades, the quality of learning data is critically important for improving AI accuracy. At Roxy AI, we enhance accuracy by organizing the learning data. Roxy AI has many unique features, such as: - Automatic extraction of potentially inappropriate data (during/after learning) - Highly efficient annotation that is error-proof for anyone - One-click deletion/replacement of inappropriate data - One-click augmentation of missing data - Extraction of inconsistencies between learning data and test data There are many other innovations as well. Please download the PDF to see the details.

• Other inspection equipment and devices

Additional inspection of defects that are difficult to detect with rule-based methods using Roxy AI.

KEYENCE products are utilized in every conceivable field. By simply adding Roxy AI, the possibilities for inspection expand. The combination of KEYENCE products and Roxy AI leverages the strengths of both. ■ Upgrading existing inspection systems - Add AI inspection to inspection systems composed of CV-X and XG-X without coding. * Compatible with CV and XG series as well. - By using AI to re-inspect areas detected through image inspection, it is also possible to reduce false positives. ■ Utilizing KEYENCE's rich lighting and image processing capabilities for AI inspection - LumiTrax can instantly capture various images, such as specular and diffuse reflection images. It is possible to simultaneously learn and inspect multiple images taken with LumiTrax. - Supports multi-spectral images using infrared and ultraviolet colors. ■ Flexibly accommodating high-speed inspections and complex systems - Achieve high-speed inspections by combining with the VJ series cameras for machine vision. - Seamlessly integrate with KEYENCE's KV series PLC without coding. ■ Suitable for inspecting small, difficult-to-capture irregularities - Compatible with the LJ series, which measures height information using lasers.

• PLC

High speed and high precision are a given! The appearance inspection software 'Roxy AI' that can be expertly used on-site.

Roxy AI is an AI inspection product that can be effectively utilized on-site. Roxy AI is not a "multifunctional" product, but rather aims to be "simple and high-performance." By limiting its functions and leveraging unique technologies that are optimal for those functions, it achieves high precision and speed. Additionally, by narrowing down its features, it has become a product that anyone can easily use. ■ Create high-precision AI inspections on-site Roxy AI organizes data (images of the inspection targets) to train the AI. The person who understands the data best can create the best AI. ■ No AI expertise required With overwhelmingly simple one-click annotation, visualization of learning progress, and a step-by-step AI creation process that reliably improves accuracy, it is packed with unique features that anyone can easily use. ■ Applicable for a wide range of uses It achieves high precision not only in defect inspections but also in detecting foreign substances and residues. It is suitable for various industries, including not just manufacturing but also food products and lumber.

• Other inspection equipment and devices

As shown in the sample video, Roxy AI is highly accurate. Here, we introduce the concept of Roxy AI, which enables high-precision inspections.

- A simple AI focused on necessary functions It can only detect defects, identify types of defects, and detect unknown defects. It cannot measure the size of defects, nor does it have functions like OCR. High accuracy is achieved with a proprietary AI optimized for simple tasks. - Learning features of both normal and defective products By learning both normal and defective products together, it accurately understands the boundaries. Learning only with normal products is appealing to customers who do not have defective products. However, to make a decision on implementation, isn't it necessary to collect a certain number of defective products? If so, I would like to improve accuracy by making a bit more effort to collect defective products. This is Roxy's thinking. - A thorough process to eliminate inappropriate training data If there are errors in the training data, the AI will learn those errors as well. By detecting and organizing inappropriate data in various ways, we prevent the creation of a distorted AI. - Annotation that eliminates mistakes regardless of who does it Annotation involves simply clicking near the defect. Even if there is a slight deviation, it does not affect accuracy. No special technique is required. All that is needed is the experience of the inspector to accurately determine whether it is a defect or not.

• Visual Inspection Equipment

Not just cost reduction! Various benefits of automated inspection.

Transitioning from visual inspection to automated inspection offers many benefits beyond cost reduction. ■ Labor Savings - Freeing valuable personnel from simple tasks and shifting them to higher value-added work - Improving the 'crowded' environment of inspection areas - Preparing for the impending labor shortage in the near future ■ Stabilization of Inspection Quality - Eliminating variations among inspectors to stabilize inspection quality - Eradicating human errors that inevitably occur - Reducing excessive detections that tend to occur as inspectors gain experience ■ Increased Operation Without Relying on Human Labor, Horizontal Expansion - Achieving 24/7 operation - Realizing horizontal and overseas expansion by AI-izing the experience of skilled inspectors ■ Improvement of Manufacturing Processes Without Relying on Memory - Accumulating inspection evidence that serves as a reference during problem occurrences The benefits mentioned here are not limited to Roxy AI, but with Roxy AI's high speed, high precision, ease of learning, and easy system integration, these benefits can be maximized.

• Visual Inspection Equipment

High speed and high precision are a given! Introducing the AI inspection product "RoxyAI" that can be effectively utilized on-site!

Roxy AI not only detects defects but also excels in foreign object contamination inspection and residue testing. It can inspect complex and diverse products and detect ambiguous defects, thereby expanding its range of applications. ■ Defect Detection - AI automatically learns judgment methods that cannot be standardized by humans. - No problem even if there are multiple types of defects such as dents, scratches, and line marks. - Any material can be inspected as long as it can be photographed, including metal, resin, and glass. - Capable of detecting not only learned defects but also unknown defects. ■ Foreign Object Contamination Detection - High-precision inspection of complex and diverse products that are difficult to assess with rule-based methods. - Detection of metal fragments, hair, insects, etc., that may be mixed in food and agricultural products. - Inspection to ensure that no similar-shaped foreign products are mixed in. ■ Residue Detection - No problem even with a lot of oil, dust, or water droplets. - Reduces the risk of defect outflow using the function to detect unknown defects. - Checks the normality of production equipment such as conveyors and verifies the cleanliness of dishes. Here are some examples of what can be achieved! Please feel free to contact us for more details.

• Visual Inspection Equipment

Fukuda Co., Ltd. has been pursuing leak detection for 60 years and is committed to advancing the future of leak testing. 【 Supporting Various Leak Tests: Air Leak Test / Gas Leak Test (Hydrogen Gas, Helium Gas) 】 We develop, manufacture, and sell leak testing equipment to inspect airtightness and sealing properties. We will propose the appropriate testers and devices for each inspection target. 【 Target Industries and Examples of Inspection Targets 】 - Automotive Industry: Engines, FC components, valves and piping, various parts - Electronic Components Industry: Smartphones, keyless switches, various small sensors - Pharmaceuticals / Food / Cosmetics Industry: Bottle containers, syringes, vials, eye drop containers, pillow packaging, PTP packaging We will propose the most suitable leak testers for each industry. 【 ISO Certification 】JIS Q 9001:2015 (ISO 9001:2015)

This time, the ACS-5000 is available as a facial recognition terminal compatible with FaceMe Security's add-on feature "Face Terminal," enabling access control management that combines "advanced anti-spoofing features" with high-speed facial recognition powered by edge AI. ■ Main Features 1. Accelerated Authentication Processing FaceMe Security is optimized for the MediaTek Genio 700's NPU, achieving faster facial recognition processing and reduced CPU load compared to processing solely on the CPU. It ensures stable authentication without delays even during peak times such as clocking in and out, contributing to alleviating congestion. 2. Advanced "Anti-Spoofing" Optimized for the IR + RGB camera mounted on the ACS-5000, it can quickly detect fraudulent authentication using photos, videos, or 3D masks within 0.6 seconds, ensuring reliable identity verification. 3. Door Control Without Additional Equipment FaceMe Security can directly control the digital I/O of the ACS-5000, allowing for the construction of an access control management system without the need for additional equipment. This simplifies system configuration and reduces implementation costs.

The non-contact beam touch sensor "HA-T401/HA-T520" has a compact detection range, making it an ideal automatic door sensor for installation on automatic doors facing narrow corridors or busy streets. Depending on the application and installation location, you can choose between "non-vision mounting type" and "non-vision built-in type." 【Features】 ● With the beam touch sensor, you can open and close the automatic door simply by bringing your hand close without touching it, ensuring hygiene. ● Depending on the installation environment, such as single sliding or bi-parting doors, the detection range can be set in four ways: left, center, right, with 4 spots and 12 spots settings. ● Two types of activation row settings are available to reduce unnecessary opening and closing of the automatic door due to cross traffic, contributing to energy-saving effects. ● The touch switch and infrared sensor are integrated, allowing for a switch between beam touch and infrared sensor modes with a single model. 【Recommended for such places!】 ☑ Entrances facing busy corridors ☑ Counters or signs near the door ☑ Buildings facing narrow corridors ☑ Buildings that prioritize aesthetics ☑ Hospitals and facilities that consider hygiene ☑ Large facilities with many doors ◎ For more details, please download the materials or contact us.

Introducing a "non-destructive testing device as an alternative to the submerged test" for pharmaceutical packaging products. "Complete vial leak testing device | MSQ-2002 series" Fully automating leak tests for various vials, bottles, eye drops, and cylindrical containers. Please check the link below for videos and catalogs.

This is an air leak tester for battery cases of electric vehicles (EVs). The battery case has the characteristic of being "large with a significant internal volume, and its structure may deform due to internal pressure." As a result, air leak testing generates "noise caused by its large size and capacity" and "noise caused by expansion or deformation due to internal pressurization." To mitigate this noise, this device features a newly developed smoothing function and a linear fitting correction function used in combination during measurement. Additionally, to achieve low-pressure control that is difficult to manage and to expand flow for large volumes, we provide a bypass box (CBU-600) equipped with an electro-pneumatic regulator for precise pressure control and a pressurization valve with a large effective diameter to supply high flow rates. Measurement methods: Work-Work comparison, Work-Master comparison, Fixed Master comparison Measurement pressure range: 10 kPa, 20 kPa, 50 kPa (depending on specifications) Gauge pressure sensor accuracy: ±2% F.S. Differential pressure sensor accuracy: ±0.5% F.S. Group settings: 0 to 31 groups Power supply voltage / Power consumption: AC100 to 240 V / 100 VA