スカイロジック List of Products and Services

This time, we used a PTZ camera with the aim of capturing multiple locations with a single camera. A PTZ camera is one that has the following mechanisms: Pan: to move the camera horizontally Tilt: to move the camera vertically Zoom: to zoom in (optically changing the angle of view). (Images are a general representation of PTZ cameras.) Let's quickly check the field of view. *For more details, please refer to the related link (blog).

• Surveillance Cameras

Recently, there has been an abundance of inexpensive network cameras available, which has allowed our EasyMonitoring2 system (a system that collects images from network cameras for tasks such as meter reading) to be implemented at a lower cost. However, this has also brought about certain issues. Specifically: - Most cameras have become of a type that does not allow lens replacement. - The predominant lenses are those designed to cover a wide area, which often results in significant image distortion (similar to fisheye lenses). Due to these trends, it has become challenging to capture images of meters from a distance using inexpensive cameras. Since the EasyMonitoring2 system connects to a large number of cameras, there is a demand for compact and affordable options. Therefore, we decided to test a method that allows for capturing images of meters even with low-cost cameras. *For more details, please refer to the related link (blog).

• Surveillance Cameras

In conventional image processing, it is possible to inspect the degree of terminal insertion, but there are challenges such as limited connector shapes and the need for positioning. Therefore, we decided to use AI for the inspection. By training the AI to recognize uninserted, partially inserted, and fully inserted states, it becomes possible to identify them. While the AI object recognition function is sufficient for "detection," the real issue is how to make the judgment. AI object recognition establishes inspection by having humans create settings that search for and recognize defects, such as scratches or dents, from the overall workpiece—if even one defect is found, it is deemed a failure (NG). In other words, object recognition alone can identify an object, but it cannot make a judgment that it is acceptable just because it appears in a certain alignment. This is where the OCR function comes into play. *For more details, please refer to the related link (blog).

• Image Processing Software

Recently, we released a new product, the "AI Sheet Counter ai-Numbers." ai-Numbers is a new counting machine that uses AI to handle sheet materials where counting has been difficult due to variations in cross-sections caused by the state of cutting blades and significant differences in thickness and appearance based on the material. Today, we will introduce the differences between the counting machine "ai-Numbers" and "EasyNumbers." In manufacturing, "counting errors" directly lead to reduced yield and complaints. Image processing-based counting machines can resolve this issue at once, but the optimal solution varies depending on whether the edges of the counting targets are complex and difficult to understand or if the focus is on counting visually clear edges as quickly as possible. We will compare our counting machines, ai-Numbers and EasyNumbers, in terms of technology, application areas, and performance, and summarize the key points for selection. *For more details, please refer to the related link (blog).

• Counting Machine

It can be combined with rule-based functions. When trying to measure the area or dimensions of a target object within a complex background, the background interfered, making it difficult to detect edges as desired. However, by extracting the target object and binarizing it using AI segmentation, inspections have become much easier. *For more details, please see the related link (blog).

• Image Processing Software

We will introduce common challenges heard from customers and ways to solve them. Challenge 1) It is difficult to analyze why a judgment was incorrect. Challenge 2) I want to quantitatively determine whether the completed model is good or bad. Challenge 3) When learning, I want to visually know which items are likely to be confused. Challenge 4) I don't know when to stop learning. When does overfitting begin? Challenge 5) Defective products are not being collected for new products. *For more details, please see the related link (blog).

• Image Processing Software

We will introduce settings that can shorten inspection time. *For more details, please see the related link (blog).

• Image Processing Software

When implementing an image recognition and inspection system using AI (deep learning), we often encounter the question, "How much training data is needed?" To cut to the chase, the answer is "there is no fixed number," but the key to successful implementation lies in "using good data appropriately." This article explains the significance and utilization of training data as follows: - The roles of training data, validation data, and evaluation data - Designing to prevent data bias - The required amount of data depends on "variability factors" - Label accuracy and annotation design - Ongoing maintenance after operation *For more details, please see the related link (blog).

• Image Processing Software

One common question from companies considering the introduction of image recognition and inspection using deep learning is, "How many images do we need to prepare for inspection?" The conclusion is that it cannot be definitively stated. This is because the amount of data required varies significantly depending on the complexity of the object's appearance (color, shape, angle, etc.) and the changing features. In this article, we have organized the basic verification process as follows: 1. Capture images of the product to be inspected and collect approximately a few hundred images (for example, around 200). 2. Select half of those images and provide "annotations" for the object. 3. Use the annotated data as training data for the AI and validate it with the remaining images (testing for recognition). 4. If there are misrecognitions or missed recognitions, increase the number of images or review the annotations and retry. 5. Repeat this "data augmentation → training → validation → readjustment" process until satisfactory accuracy is achieved. *For more details, please refer to the related link (blog).

• Image Processing Software

NumberVision utilizes AI technology to instantly recognize and digitize vehicle license plates. This enables various applications such as parking management, enhanced facility security, customer management, and traffic statistics. It can recognize plates with 99.9% accuracy even in low light or bad weather, and since it is a one-time purchase model, there are no ongoing costs. It can be flexibly utilized according to the customer's environment, allowing for license plate recognition in diverse scenarios. 1) Capture vehicles with your existing network camera. 2) NumberVision recognizes the vehicle license plate → digitizes and records the data. 3) Management can be tailored to your needs. ■Features 【Catering to Various Needs】 Can be utilized in various settings, including parking lots, factories, commercial facilities, hospitals, and gas stations, both indoors and outdoors. 【Easy and Quick Implementation】 All necessary software, PC, and cameras are provided, allowing for immediate operation after purchase, and it is a one-time purchase model that eliminates troublesome operational costs. 【Overwhelming Recognition Accuracy】 Recognizes license plates with 99.9% accuracy regardless of day or night, even in low light or at dusk.

• Image Processing Software

The AI general-purpose appearance inspection software "EasyInspector2" has newly added the "AI Segmentation" feature. The "AI Segmentation" feature performs segmentation (region division and labeling of each pixel in the image) using deep learning within the image. It calculates the ratio of pixels detected by segmentation within a specified polygonal area and makes pass/fail judgments. This feature can mainly be used for the following applications: - Tank water level - Area of metal processing surfaces and coating area - Measurement of crop growth (size) - Detection of lesions in photos and measurement of their area - Checking for coating gaps in sealants and adhesives - Monitoring the amount of smoke emitted By using the AI Segmentation feature for judgment, it achieves: - Improved work efficiency: Significant time savings compared to visual inspection - Enhanced accuracy in quality control: Reducing human errors and stabilizing quality - Cost reduction: Lower labor costs and reduced costs due to suppression of defective products.

• Image Processing Software

During the inspection stage of wood cutting, you were considering cutting a board measuring 2m in length and a certain width from 5mm to 160mm while avoiding defective parts and cutting at the right timing. 【Inspection Settings and Results】 For knots, we used the "Color Comparison Inspection / Presence or Absence of Specified Color" function of EasyInspector. The specified color was set to the color of the knots. An inspection frame the size of one kamaboko board was set from the edge of the board, and this area was divided into four parts for the presence or absence color inspection. To capture the roughness of the board, we dimmed the indoor lighting and illuminated it from an angle, which allowed us to detect the roughness. The inspection item was "Scratch Inspection." No alignment correction was applied for the inspection of the flowing boards.

• Image Processing Software

This was an inquiry from a trading company interested in our software EasyInspector, asking whether standard features could accommodate their needs, and if not, whether customization would be possible. 【Inspection Settings and Results】 By using the "Dimension Angle Inspection" feature of EasyInspector, we measured two dimensions and made a judgment in less than one second. Inspection frame 001 measured the inner diameter of a circle (251.50 pixels), and inspection frame 002 measured the black width (59.50 pixels). In this verification, measurements were possible with standard features, and while customization can be accommodated according to customer requests, it will incur additional costs. 【Software Used】 Software Used: EasyInspector200 Number of Inspection Points: 2

• Image Processing Software

The packaging design of food products often changes in small parts or colors during campaigns, and there have been instances of mistakenly shipping packaging due to similar designs, which has caused some trouble. This time, we determined the differences in design and the differences in the same design materials. 【Inspection Settings and Results】 By using the "Comparison with Master Image" feature of EasyInspector, we detected design differences in the entire imaging screen frame and were able to identify visually similar products (different items) in under 50 seconds. 【Software and Equipment Used】 Software Used: EasyInspector710 Field of View: A4 Size Minimum Size of Inspection Target: 5mm Number of Inspection Points: 1 Camera Resolution: Our Company Scanner Current 'EasyInspector2' color package can be inspected using the "Comparison with Master Image" feature.

• Image Processing Software

It seems that the inspection involving a mix of letters and numbers was difficult to determine at the time. With improvements in lighting and settings, we were able to read with higher accuracy. This request was made in 2015, but the OCR reading function in the current EasyInspector has been enhanced, increasing its accuracy, allowing us to retain images and record content in a CSV file. 【Inspection Settings and Results】 By using EasyInspector's "Character Recognition (OCR)" function, we detected readings from 16 printed locations, achieving a probability accuracy of about 90%, and we were able to make determinations in under 3.41 seconds. A 3-megapixel camera provided the most stable inspections. It is indeed easier to misdetect when letters and numbers are mixed; for example, "8" and "B" are characters that are prone to misdetection. However, when "8" is among the numbers and "B" is among the letters, we were able to read them relatively accurately and stably. Although the number of inspection frames increases, we believe that separating letters and numbers and setting inspection frames accordingly will enable stable inspections.

• Image Processing Software

It has been reported that there was an incident where products with similar package types were shipped with incorrect labels. In many cases where multiple types are manufactured in a flow production system, visual inspection systems play a crucial role. 【Inspection Settings and Results】 By using the "Comparison with Master Image" feature of EasyInspector, judgments could be made in less than one second. The color judgment tolerance was set to 69. As shown in the image above on the right, differences from the master image are detected in red. 【Software Used】 Software Used: EasyInspector310 Number of Inspection Points: 1 The current 'EasyInspector2' color package can be inspected using the "Comparison with Master Image" feature.

• Image Processing Software

This is an inspection where only the model number differs on the product label. Minor differences are often overlooked during visual checks. Inspections related to labels often involve reading and recording serial numbers using OCR functions, but there are also many operations that check whether certain parts, such as model numbers, are correct. This time, the inspection was to verify that the model number and cautionary notes were not incorrect, based on the images sent via email. 【Inspection Settings and Results】 By using the "Presence of Specified Color Inspection" feature of EasyInspector, we were able to detect differences in two locations and determine whether the visually similar label markings were correct in less than 0.10 seconds. 【Software Used】 Software Used: EasyInspector300 Inspection Locations: 2 The current 'EasyInspector2' color package can perform inspections for the "Presence of Specified Color."

• Image Processing Software

This is a judgment test based on the submitted images. If the inspection target can be captured clearly, it can generally be said that inspection is possible. 【Inspection Settings and Results】 By using the "Dimension Angle Inspection" feature of EasyInspector, we were able to measure the angle at one location and make a judgment in 0.09 seconds. 【Software Used】 Software Used: EasyInspector310

• Image Processing Software

**Inspection Settings and Results** By using EasyInspector's "Comparison with Master Image" feature, inspections for four areas of burrs, chips, and shapes were possible. Backlight illumination was used to enhance the contrast between the product and the background for inspection. **Software and Equipment Used** Software Used: EasyInspector310 Field of View: 200 x 150mm Minimum Size of Inspection Target: 0.5mm Number of Inspection Points: 4 Camera Resolution: 1.3 Megapixels Lens Focal Length: 35mm Distance Between Lens and Product: 130mm Lighting: Backlight Current 'EasyInspector2' color package allows inspection with "Comparison with Master Image."

• Image Processing Software

Currently, cases that are verified with DeepSky were previously verified with EasyInspector before the release of DeepSky. This verification involves inspecting the shape of the base of green onions. It includes sorting those with a bulbous, rounded base and many roots. The customer has specific requirements for the camera and resolution. 【Inspection Settings and Results】 We are inspecting the shape of the base of green onions using EasyInspector's "Dimension and Angle Inspection" function and "Damage Inspection" function. We propose various measurement methods across 14 different items. The inspection content is something that our software DeepSky, released in 2020, excels at. Our company aims to solve customer issues as cost-effectively as possible by utilizing the equipment they currently have, such as cameras.

• Image Processing Software

At the request of an LED lighting manufacturer, we conducted a verification test for dirt detection. This test aims to detect dirt such as film-like dust that is difficult to see with the naked eye. 【Inspection Settings and Results】 We received four types of samples and were able to detect three types of dirt. However, for the sample with a bumpy surface finish, we were unable to detect dirt under the same imaging conditions. How to capture the defective areas becomes a crucial point in the inspection. The inspection frame was set as a single frame for the entire area, with an inspection cycle of 2.03 seconds. 【Software and Equipment Used】 Software used: EasyInspector710 Field of view: 64 x 51 mm Minimum size of inspection target: 0.2 mm Number of inspection points: 1 Camera resolution: 10 million pixels Lens focal length: 25 mm for high pixels Distance between lens and product: 260 mm Lighting: LED lighting (illuminated from both sides)

• Image Processing Software

Sintering with silver paste causes less damage to the materials compared to soldering, and it can be said that this material will be used more widely in the future as a substitute for solder. 【Inspection Settings and Results】 By using the "Scratch and Defect Inspection" feature of EasyInspector, we were able to detect both chipping and smudging in 0.40 seconds by applying one inspection frame to the entire area. The size (resolution) per pixel is approximately 7μm (0.007mm). In actual operation, our "sm＠rtROBO" is used to move the inspection area while being inspected by a single camera. 【Software and Equipment Used】 Software Used: EasyInspector Field of View: 14 mm Minimum Size of Inspection Target: Approximately 7μm (0.007mm) Number of Inspection Points: 1 Camera Resolution: 5 million pixels Lens Focal Length: Not recorded Distance Between Lens and Product: 95mm Lighting: Backlight illumination The current 'EasyInspector2' color package can be used for [Scratch and Defect Detection].

• Image Processing Software

In the label manufacturing industry, such as for product descriptions, productivity is increased by printing multiple labels at once. On the other hand, issues such as missing prints, smudges, and defects due to black spots have also arisen during printing. This problem of defective shipments is one that many label manufacturing companies face, and image processing software has been used as a means of improvement. 【Inspection Settings and Results】 By using the "Comparison with Master Image" feature of EasyInspector, three inspection frames were set for a single label to establish pass/fail criteria for missing prints, smudges, and black spots, detecting multiple defective areas. In the image above, the smudged area (NG) is highlighted in red, while the OK areas remain unchanged. 【Software and Equipment Used】 Software Used: EasyInspector710 Field of View: 60x 70mm Minimum Size of Inspection Target: Approximately 5mm Number of Inspection Locations: 3 Camera Resolution: 1.3 Megapixels Lens Focal Length: 35mm Distance Between Lens and Product: Approximately 360mm Lighting: Indoor Lighting Current inspections can be performed with the 'EasyInspector2' color package using the "Comparison with Master Image" feature.

• Image Processing Software

In the label manufacturing industry for product descriptions, there are many cases where numbers need to be printed. Even if we can detect smudges or black spots, it is meaningless if the printed numbers themselves are incorrect. Therefore, we considered introducing a function to recognize numbers to detect "character discrepancies" and "printing errors." 【Inspection Settings and Results】 By using the "OCR Pro" feature of EasyInspector, we can read multiple lot numbers (corresponding to the number of characters) from a single inspection frame. For characters that are difficult to read, it is also possible to train the system to inspect them one by one. The image on the right above shows that the number "8306" has been successfully read. 【Software and Equipment Used】 Software Used: EasyInspector710 Field of View: 60 x 70 mm Minimum Size of Inspection Target: Approximately 5 mm Number of Inspection Points: 1 Camera Resolution: 1.3 Megapixels Lens Focal Length: 35 mm Distance Between Lens and Product: Approximately 360 mm Lighting: Indoor Lighting The current 'EasyInspector2' RD (ReaDing) package can perform inspections with [OCR (Character Recognition)] and [Machine Learning OCR].

• Image Processing Software

The numbers and alphabetic characters displayed on digital meters used by electronic and electrical equipment manufacturers are mostly in a 7-segment format. There was a need for software that could read those numbers and characters within a PC. 【Inspection Settings and Results】 Using the "OCR Pro" feature of EasyInspector, we trained the three alphabetic characters "C, H, K" one by one. Due to the limited display format, there were instances where similar characters were misrecognized, but by adjusting the sensitivity and retraining, we were able to correctly recognize the characters. 【Software and Equipment Used】 Software Used: EasyInspector710 Field of View: 160mm x 200mm Minimum Size of Inspection Target: Approximately 30mm x 50mm per character Number of Inspection Locations: 3 Camera Resolution: 1.3 Megapixels Lens Focal Length: 6mm Distance Between Lens and Product: Approximately 180mm Lighting: No record The current 'EasyInspector2' RD (ReaDing) package can perform inspections with [OCR (Optical Character Recognition)] and [Machine Learning OCR].

• Image Processing Software

When inspecting automotive parts after painting, there are instances where defects caused by painting errors result in the shipment of defective products. We wanted to enable the detection of these defects through image inspection. In automotive parts manufacturing, post-painting defect inspection is considered crucial for product assembly, leading us to consider the introduction of image processing to ensure accuracy. 【Inspection Setup and Results】 Using the "Scratch and Defect Inspection" feature of EasyInspector, we set up inspection frames to detect bright pixels compared to surrounding pixels and dark pixels at the same position, allowing detection regardless of which type is present. Even in cases where the background has a striped pattern, we were able to detect defects as small as 1mm by modifying the settings.

• Image Processing Software

The stickers applied during the manufacturing process of automobile manufacturers can lead to delays in shipping if they are forgotten or misapplied. Therefore, it was suggested that image processing might be more efficient than visual inspection, and we received a request from our company. 【Inspection Settings and Results】 This was verified by sending images. Using the "Presence of Specified Color Inspection" feature of EasyInspector's "Color Comparison Inspection," we verified whether the color of the sticker was present within two inspection frames. We specified the inspection frames to ensure that the upper and lower stickers did not overlap, and we were able to detect the specified color within those frames. 【Software Used】 Software Used: EasyInspector310 Minimum Size of Inspection Target: Approximately 30mm x 30mm Number of Inspection Locations: 2

• Image Processing Software

In the automobile manufacturing process, buttons are installed, and when pressed, the operation begins. However, if the installation order is incorrect, the marks and operations may not match, leading to the possibility of shipping defective products. Such defects can result in significant losses for automobile manufacturers, so an image processing system was used to enhance accuracy. 【Inspection Settings and Results】 Using the "Color Comparison Inspection" feature of EasyInspector, specifically the "Comparison with Master Image," inspection frames were designated for the marks of four buttons to check if the mark images matched. The white marks were successfully identified against a black background. 【Software and Equipment Used】 Software Used: EasyInspector310 Field of View: 125mm x 170mm Minimum Size of Inspection Target: Approximately 20mm x 30mm Number of Inspection Points: 4 Camera Resolution: 1.3 Megapixels Lens Focal Length: 6mm Distance Between Lens and Product: Approximately 200mm Lighting: Not used The current 'EasyInspector2' color package allows for inspection using the "Comparison with Master Image" feature.

• Image Processing Software

In the automotive manufacturing industry, the panels to be installed have many holes, and because there are similar panels, if there is any issue with the presence or position of even one hole, the product cannot be deemed acceptable. Therefore, we considered using image recognition technology. [Inspection Settings and Results] We included a panel with two large holes and 27 small holes within the field of view. For the larger holes, we set up the EasyInspector's "Color Comparison Inspection" feature to detect the "pink color" of the holes. For the smaller holes, we used the "Defect Inspection" feature to detect white defects. The holes themselves were clearly captured using backlight illumination, allowing for accurate inspection even if the position of the holes was off by 50mm. The red areas on the right side of the image indicate where the holes were detected.

• Image Processing Software

It is common to use heat-resistant ceramic sheets in situations where products are handled. They come in various thicknesses, ranging from about 60μm for thin sheets to about 200μm for thicker ones. This time, we solved the problem of measuring the number of 200μm ceramic sheets handled by an electronic component manufacturer, which was previously done manually one by one, by using our software that allows for non-contact measurement of the sheets. 【Inspection Settings and Results】 The verification was done using only photographs. By using EasyInspector's "Luminance Change Inspection," we were able to distinguish between light and dark, and we captured images with 23 thick sheets stacked. Shadows were created between the stacked sheets, making the contrast between light and dark clear, and due to the consistency of the bright and dark lines, we were able to conduct the inspection smoothly. 【Software Used】 Software Used: EasyInspector310 Field of View: 10mm x 13mm Minimum Size of Inspection Target: Approximately 0.2mm (200μm) Number of Inspection Points: 23 Measurement of the Number of Ceramic Sheets (Thin) https://skylogiq.co.jp/DIY_HowTo/860

• Image Processing Software

When measuring the number of approximately 60μm thin ceramic sheets handled by an electronics manufacturer, there were disadvantages such as counting errors and the time taken for visual inspection. Therefore, we considered using our image processing software to enable non-contact counting of the sheets. 【Inspection Settings and Results】 Using the "Luminance Change Inspection" function of EasyInspector, we conducted verification by stacking 63 ceramic sheets of 60μm size. If we can clearly show the difference in brightness, inspection is possible. By uniformly illuminating from the front and aligning the product bundle evenly, we were able to count the number of sheets of 60μm size. However, there was a challenge in the stage of setting the inspection frame, as we had to avoid areas with dust and other contaminants. 【Software Used】 Software Used: EasyInspector310 Field of View: 9mm x 12mm Minimum Size of Inspection Target: Approximately 0.06mm (60μm) Number of Inspection Points: 63 Counting of Ceramic Sheets (Thick) https://skylogiq.co.jp/DIY_HowTo/856

• Image Processing Software

There are tasks involving the application of logo marks from various manufacturers. It is a common mistake to accidentally apply rectangular or square logo marks in reverse. Our company supports you with image inspection to ensure that your important logo is shipped with high design quality and beauty. 【Inspection Settings and Results】 By using the "Comparison with Master Image" feature of EasyInspector, we were able to determine a labeling error in less than 3.56 seconds. The "OK" and "NG" images provided had differences in field of view, brightness, and brightness direction, so we adjusted the "Shift Correction" feature and the settings for "Color Judgment Tolerance Range." It is necessary to prepare the inspection environment during actual operation. (Fixing camera position and orientation, fixing lighting position and intensity, fixing inspection item placement, etc.) 【Software Used】 Software Used: EasyInspector710 The current 'EasyInspector2' color package can perform inspections using the "Comparison with Master Image" feature.

• Image Processing Software

This is an inspection to find black spots on the surface of the customer's product, requested by the trading company. They wanted to detect black spots across the entire screen, so we received images and conducted a judgment test. They requested information (coordinates) indicating where the foreign substances are located. 【Inspection Settings and Results】 Detection was achieved using the "Scratch and Defect Inspection" feature of EasyInspector. The screen was divided into 30 sections, creating 30 inspection frames. It is possible to keep records of which inspection frame detected defects, along with images and CSV files. The test was conducted with strict detection settings, so the customer will need to set their own criteria for what constitutes a defect. Our company offers affordable sales by having customers set the parameters themselves. 【Software Used】 Software used: EasyInspector710 The current 'EasyInspector2' color package can be used for inspection with the [Scratch and Defect Detection] feature.

• Image Processing Software

At the exhibition, we conducted sample verification of precise measurements of CDs at the request of a trading company that actually tried our inspection software. 【Inspection Settings and Results】 By using the "Dimension and Angle Inspection" feature of EasyInspector, measurements of approximately 62 to 73 microns were achieved. This time, a high-resolution camera with a wide field of view was used. The inspection time was about 0.8 seconds. By narrowing the field of view and reducing the resolution, the inspection time can be shortened. The green thin line in the left image indicates the measured part that was detected. In the right image, we have submitted the evaluation of repeatability accuracy for five repetitions at one location. 【Software and Equipment Used】 Software Used: EasyInspector710 Field of View: Not recorded Minimum Size of Inspection Target: 0.01mm (10 micrometers) Number of Inspection Points: 2 Camera Resolution: 5 million pixels Lens Focal Length: Magnification x0.7-x4.5, Working Distance 52mm, Lens Holder Diameter φ50mm, Total Lens Length 190mm Distance Between Lens and Product: 300mm Lighting: Indoor light The current 'EasyInspector2' MS (MeaSure) package can perform inspections for [Position and Width Measurement] and [Angle Measurement].

• Image Processing Software

The instrument manufacturer was experiencing issues with incorrect placement of knobs and dials in their audio products. It is conceivable that human error could occur during visual inspections. They were considering software that could simultaneously inspect multiple factors, such as color differences and incorrect orientations of the knobs. 【Inspection Settings and Results】 Using seven types of image data provided, we were able to detect insertion errors, etc., within our company using the "Master Image Comparison" and "Presence of Specified Color Inspection" functions of "EasyInspector." The inspection time for each part ranged from 0.90 seconds to 0.58 seconds. 【Software Used】 Software Used: EasyInspector310 Number of Inspection Points: 2 to 10 points for each of the 7 images The current 'EasyInspector2' color package can perform inspections using the [Presence of Specified Color] and [Master Image Comparison] features.

• Image Processing Software

Some production technology personnel believe that it is difficult to inspect work with high visibility, such as glass, due to halation. Our company also proposes inspection methods using cameras, lenses, and lighting that can be used for photography. 【Inspection Settings and Results】 By using the "Luminance Change Inspection" function of EasyInspector, we were able to verify the number of stacked glass sheets and make a judgment in 0.04 seconds. We detected the black areas and accurately counted the number of sheets. However, we could not make accurate judgments (detect differences in brightness) in areas where light was not clearly shining, making lighting an important factor in this case as well. 【Software Used】 Software Used: EasyInspector300 Number of Inspection Points: 1

• Image Processing Software

Children have loved card games both now and in the past. Many of you may have collected cards when you were young. Some printing manufacturers produce cards specifically for card games. Once packaged, it becomes impossible to know which cards are inside. Therefore, we utilize our EasyInspector to keep a record of the issued cards before packaging. Inspection settings and results By using the "OCR Pro" feature of EasyInspector, we were able to read a single location with a 7-character format (card number) in 0.19 seconds. The OCR Pro feature includes binarization and dictionary learning functions. It can read not only existing fonts but also other types, although it may become difficult to read if the lines of the characters are broken or if adjacent characters are connected. Please contact us for inquiries about the extent of readability.

• Image Processing Software

There was a request for evaluation of image inspection in multiple areas to reduce personnel and improve operational efficiency in the cabinet manufacturing process. 【Inspection Settings and Results】 By using EasyInspector's "Scratch Detection" function and "Presence of Specified Color Inspection" function, we were able to detect 20 differences and determine visually similar items (different products) for each category in less than one second. 【Software and Equipment Used】 Software Used: EasyInspector710 Field of View: Various field of views Minimum Size of Inspection Target: 2mm Number of Inspection Points: 20 Camera Resolution: 5 Megapixels GigE Lens Focal Length: 25mm Distance from Lens to Product: 1200mm Lighting: Linear fluorescent lights, etc. Distance from Lighting to Inspection Item: Not recorded The current 'EasyInspector2' color package can inspect for the presence of specified colors.

• Image Processing Software

In the food industry, foreign object contamination can lead to significant accidents. As of 2017, verification was conducted using the conventional inspection software EasyInspector, but now such cases can be easily configured and inspected with DeepSky. Please refer to the article "Stopping the Conveyor After Discovering Insects on Cabbage." 【Inspection Settings and Results】 By using EasyInspector's "Presence of Specified Color Inspection" feature, it was possible to detect different colors from one location (the entire field of view). However, since the inspection was based on submitted images, there were instances where the background gray was similar to the different color, leading to noticeable false detections. While it is possible to set a strict "color judgment tolerance range," it has been suggested that adjustments to background color and lighting are necessary. 【Software and Equipment Used】 Software Used: EasyInspector710 Number of Inspection Points: 1 (entirety) Current 'EasyInspector2' color package allows inspection for the "Presence of Specified Color."

• Image Processing Software

This is an inquiry from overseas. We have partnered retailers in China and Malaysia. 【Inspection Settings and Results】 By using the "Scratch Inspection" feature of EasyInspector, we were able to determine one location (with a total of two inspection frames) in 0.07 seconds. We applied two types of inspection frames for black and white spots. For detecting even finer fraying, it may be possible to improve detection by increasing the camera resolution or adjusting the lighting. 【Software Used】 Software used: EasyInspector710 Number of inspection locations: 1 (overall)

• Image Processing Software

In convenience stores and dessert shops, it has become common practice to sell items with cutlery included. This time, we will inspect small transparent spoons that are packaged in transparent film on top of bubble wrap. Our company is committed to creating an environment tailored to our customers' inspection situations, providing free evaluations and reports. 【Inspection Settings and Results】 By using DeepSky's inspection features, we easily set up a configuration where if one spoon is present in the screen area, it displays "OK." On the left is the annotation image (the specified image used to teach what to detect), and on the right is the image detecting the spoon. DeepSky can make judgments without fixed positioning, even in situations where it is difficult for a person to distinguish by visual confirmation. 【Software and Equipment Used】 Software Used: DeepSky Field of View: 150 x 120mm Minimum Size of Inspection Target: 70mm Number of Inspection Points: 1 (configured to display "OK" when one spoon is detected and "NG" when no spoon is detected; one label) Camera Resolution: 1.3 million pixels Lens Focal Length: 8mm Distance Between Lens and Product: 430mm Lighting: Indoor fluorescent lights

• Image Processing Software

The seat designs of automobiles, buses, airplanes, and other vehicles may vary in design and size due to different specifications, and the internal components also change depending on the grade. In the case of similar products, accidents occurred where similar but different items were shipped because they looked almost identical. 【Inspection Settings and Results】 An evaluation was conducted using DeepSky. On the left is the annotation image (the teacher image that teaches the part to be detected), and on the right is the inspection result image. This time, since it was a work that could not be externally leaked, our company used a model and proposed an annotation method. Currently, for consideration of implementation, we are offering a free rental service for demo units. The number of demo units is limited, so depending on the reservation status, there may be delays. We kindly ask you to make your reservations as early as possible. 【Software and Equipment Used】 Software used: DeepSky Field of view: 100 x 80mm Minimum size of inspection target: 5mm Number of inspection points: 4 (Label 8) Camera resolution: 1.3 million pixels Lens focal length: 12mm Distance between lens and product: 450mm Lighting: Indoor fluorescent lights

• Image Processing Software

In the pharmaceutical industry, similar to other sectors, products were packed in paper boxes with a specified number and sealed with stickers for shipment. However, incidents of forgetting to apply the seals occurred. We verified this using DeepSky, which can detect the absence of seals regardless of various orientations and angles during the operation. DeepSky is a software that excels in such inspections and is utilized across various industries. 【Inspection Settings and Results】 By using DeepSky's inspection function, we were able to detect the presence or absence of a seal in one location, achieving a judgment time of 0.23 seconds. The inspection time may vary depending on the specifications of the computer. The images were extracted from the report. In the left image, we set it to detect one seal with one type of label as OK, while in the right image, the inspection judgments of "OK" and "NG" were made without false detections. The inspection was conducted with the images provided, but this time we are also considering the environment, such as cameras. The camera and lens will vary depending on the acceptable working distance and field of view.

• Image Processing Software

We received 100 sample images from a manufacturer that produces high-quality films for verification. For the free evaluation of DeepSky, it is necessary to send photos of various defective shapes of the work. 【Inspection Settings and Results】 By using DeepSky's inspection function, we were able to detect the presence of various shapes of defects with an inspection cycle of 0.28 seconds per instance. The left image shows the settings, while the right image outlines the areas we want to identify and labels them for learning purposes. - We created a setting where if any NG (defective) item is detected, it is marked as a failure; if only OK items are detected or nothing is detected, it is marked as a pass. By training the software with OK and NG, we have been able to conduct inspections that meet our customers' requirements. Our website offers a trial service for DeepSky. Please take advantage of it. 【Software Used】 Software Used: DeepSky Learning Version

• Image Processing Software

Out of the 84 sample images you sent (39 NG / 45 OK), 50 images (25 each) were used for training as teacher images. The remaining untrained images were used as test images. 【Inspection Settings and Results】 By using DeepSky's inspection function, we were able to determine the Ito-biki in 0.23 seconds. The results of inspecting the teacher images and the remaining 34 test images showed a correct answer rate of 100%. The number at the top of the image detection frame indicates the AI's confidence level in percentage (number of recognition points). 【Software Used】 Software used: DeepSky Learning Edition

• Image Processing Software