スカイロジック List of Products and Services

This is a verification request from manufacturers of automotive catalysts, focusing on automotive catalysts for passenger cars, diesel vehicles, and motorcycles. The deep learning image inspection software 'DeepSky' allows for easy collection, learning, and judgment of image data using a single PC. In this instance, the shape discrimination was performed using 'DeepSky'. This software is trained in advance to distinguish between a "set of rectangles" and a "set of hexagons" across the entire screen. It utilizes AI (Deep Learning), and the results of the verification showed that it is possible to detect and discriminate between rectangular and hexagonal shapes. Even for inspection cases that you may have previously given up on, please feel free to consult with us. [Software Used] ■DeepSky *For more details, please refer to the PDF document or feel free to contact us.

There have been verification requests from manufacturers of automotive catalysts, focusing on automotive catalysts for passenger vehicles, diesel vehicle catalysts, and motorcycle catalysts. In Hamamatsu City, Shizuoka Prefecture, where the manufacturing industry, including automotive parts, is thriving, we serve various industrial machinery and control equipment manufacturers as clients, with a track record of providing support for over 1,000 cases. We are developing inspection software that is easy to integrate and sold as a complete package. This time, the determination of clogging was performed using 'DeepSky.' By pre-training it on the "black clumps" indicative of clogging, we can detect areas within the screen that appear to be clogged. The image inspection software 'DeepSky,' which utilizes deep learning, is an all-in-one and simple-to-operate AI image inspection software themed around "easy operation for field users," and it is steadily increasing its implementation record. [Software Used] ■DeepSky *For more details, please refer to the PDF document or feel free to contact us.

The surveying equipment manufacturer who requested this verification is also a customer we have been dealing with for some time. Our company serves various industrial machinery and control equipment manufacturers in Hamamatsu City, Shizuoka Prefecture, which is active in the manufacturing of automotive parts. We have a track record of providing and supporting over 1,000 cases, and we believe that our easy-to-integrate inspection software, sold as a one-time purchase, has led to repeat business. As a result of training specifically on wall cracks, it seems difficult to completely eliminate false detections through training alone. When a target object (in this case, cracks) is found within a specific area, it is possible to decide whether or not to include it in the count through settings. However, in this case, positional deviations of the workpiece cannot be tolerated in processing, so if the workpiece is assumed to be misaligned, it may be quicker to blur the surroundings. [Software Used] ■DeepSky *For more details, please refer to the PDF document or feel free to contact us.

We would like to introduce a case of a simple inspection of sample parts received from specialized manufacturers of automotive components. The inspection was conducted using DeepSky's inspection capabilities. We were able to detect visible scratches. However, there are some scratches that were not detected, so it seems necessary to improve the lighting conditions to make them visible. On our website, we are preparing a free trial of DeepSky, which is designed for image processing in factory automation from a practical perspective, and we look forward to your visit. [Software Used] ■DeepSky *For more details, please refer to the related links or feel free to contact us.

This is a request for verification of painted parts used in the interiors of luxury cars. The increase in inquiries about "paint defects" is one example of this trend. Various paint manufacturers have adopted our image inspection technology. We offer a "simple evaluation service" free of charge, provided by our technical staff. If defects or judgments can be detected during the simple evaluation, we recommend conducting a "feasibility verification" test that simulates actual operations to evaluate processing time and judgment accuracy. Additionally, we can lend demo units free of charge, allowing users to experience the deep learning inspection software firsthand. By using DeepSky's inspection capabilities, inspections were possible. The workpieces to be inspected are conveyed on a conveyor, and one side is magnified for imaging. Adjusting the lighting when imaging painted products is a key point.

This time, I would like to focus on the production of a mechanism called "string conveyor." Unlike a regular conveyor, a string conveyor transports products along a "line" rather than a "surface." Why do we go to the trouble of supporting products with a line? The advantages are: - It allows for inspection from both the top and bottom since it can be photographed from below. - By placing a backlight underneath, it enables photography using transmitted light. - By varying the speeds of the two strings, it can rotate the products. And so on. Production time: 1.5 days Material cost: 30,000 to 40,000 yen [How to make it] This time, there will be no cutting or gluing. The basic approach is to assemble the various parts that have been arranged using a screwdriver and hex wrench. *For more details, please refer to the related links or feel free to contact us.*

Rubber molded parts, such as automotive components, have different clip mounting positions depending on the product. This time, we received a request for a simple verification of clip position measurement. If the field of view is approximately 250mm in the longitudinal direction, it seems possible to measure the clip pitch with an accuracy of ±0.5mm. By using the "Dimension and Angle Inspection" feature of EasyInspector, we were able to detect the differences in the positions of four clips and determine visually similar similar products (different items) in less than 0.93 seconds. The image on the right shows the detection frame. Our company accepts verification and support from technical staff on a daily basis. If you have any issues, questions, or uncertainties during operation, please feel free to contact us at any time. 【Software Used】 EasyInspector710 (formerly EasyInspector) The current 'EasyInspector2' MS (MeaSure) package can be used for position and width measurement inspections.

This is an inquiry regarding the monitoring of manufacturing machines from manufacturers of drive recorders and car navigation systems installed in automobiles. We have received six images for monitoring, and we are inquiring about the appropriate settings for each reading. Our company website offers a free trial of the inspection software. The suitability of the inspection software varies depending on the work, so please feel free to contact us if you have any questions. 【Inspection Settings and Results】 For three of the six images, we are using the "Specified Color Presence Inspection" function to detect the color when illuminated, and we determine pass or fail based on whether it is detected or not. For two images, we are using the "Meter Reading Function," but we believe there may be variations in the way light hits the upper and lower parts of the meter, so illuminating the lower part, which tends to be darker, with bar lighting or similar methods may reduce false detections. For one image, reading is possible with the image you provided. Some minor adjustment for misalignment is possible, but fixing the camera is essential.

A housing material manufacturer is considering our inspection software for product management of precision plastic extrusions, such as automotive parts, and has made an inquiry. 【Inspection Settings and Results】 By using the "Dimension Angle Inspection" feature of EasyInspector, we were able to measure 22 points in 0.53 seconds. For a 5-megapixel camera (with a horizontal field of view of approximately 80mm): Since the horizontal resolution is 2592 pixels, the size of 1 pixel is "80÷2592=0.0308…mm." This results in a rough inspection with increments of about 0.0308mm, but as mentioned above, judgments were possible. For a 20-megapixel camera (with a horizontal field of view of approximately 80mm): Since the horizontal resolution is 5496 pixels, the size of 1 pixel is "80÷5496=0.0145…mm." This allows for detection with double the precision compared to the 5-megapixel camera with increments of about 0.0145mm, but even with 20 megapixels, the inspection remains rough.

Painted products can be considered as challenging work for capturing images of defects. Clearly identifying defective areas is a very important factor for image inspection. We will explore different lighting techniques for verification. 【Inspection Settings and Results】 The process of enclosing the areas we want to detect in rectangles, as shown in the image, is called annotation. In practice, it is necessary to capture and annotate as many defects as possible. The number of images required depends on the inspection, but we believe around 50 images are necessary. Given the size of the defects, we propose using two cameras (considering the inspection cycle, this means two systems). By training various patterns of defects in a software called DeepSky in advance, it becomes possible to detect unknown defects as well. 【Software Used】 DeepSky

We inspected die-cast products using a software called DeepSky, which utilizes AI (Deep Learning). By training the software on the areas we want to detect, it adjusts its own settings and learns to recognize them. This time, we are assuming an operation where all items on the conveyor are inspected using DeepSky. 【Inspection Settings and Results】 This inquiry also came directly from an end user. Since our company does not sell industrial equipment like conveyors, we ask customers to purchase through a vendor. The left image shows the annotation process, where we outline the defective parts we want to identify. The right image displays the detection frames of the inspection results. The numbers indicate the confidence percentage of the AI based on the number of recognition points. 【Software and Equipment Used】 Software used: DeepSky Field of view: Approximately 46×37 mm Minimum size of inspection target: 5 mm Number of inspection points: 1 Camera resolution: 1.3 million pixels Lens focal length: 50 mm + 5 mm close-up ring Distance between lens and product: Approximately 430 mm Lighting: Indoor fluorescent lights

The request from a manufacturer engaged in metal rolling concerns a defect known as "chatter marks," which are surface defects. Sample images have been provided. Chatter marks refer to the horizontal streaks that appear at regular intervals in the direction of movement when polishing. They are commonly seen in plate polishing and are considered a significant issue for businesses that perform plate polishing, such as those working with flat bars. [Inspection Settings and Results] By using the "Scratch and Defect Inspection" feature of EasyInspector, we were able to detect chatter in one location (the entire screen) in 0.16 seconds. The setting is configured not to detect "vertical stripes." Due to the significant variations in the reflection of the light source and overall brightness caused by the lighting, it is essential to ensure that the lighting conditions are consistent during actual inspection operations. [Software Used] Software Used: EasyInspector Current 'EasyInspector2' color package [Scratch and Defect Detection] can be used for inspection.

This is a simple verification of the metal fittings of a car seat belt. Our company is located in Hamamatsu City, Shizuoka Prefecture. It is an active region for automotive parts manufacturing, and we have a track record of being adopted by over 2,000 industrial equipment manufacturers and manufacturers. We design inspection software that is easy to integrate and sold as a one-time purchase, and we have many repeat customers. 【Inspection Settings and Results】 By using the "Scratch Inspection" function of EasyInspector, we were able to inspect one location in 0.55 seconds. The image shows the imaging environment for this inspection. We verified all the sample products we received under the same lighting conditions and inspection settings. Among them, some defective products with scratches and some defective products with dacon could not be detected. Other sample products were able to detect the scratches and rust we wanted to identify.

The manufacturer of the autonomous driving assistance system has considered our character recognition function for product management. The products display multiple characters and logos. Inspections for missing or faded displays and for recording purposes are made easier with the recognition function. This is also a common case, and we have received applications from various manufacturers. [Inspection Settings and Results] As a result of verification conducted with the sample we received, it was determined using EasyInspector. For the inspection of missing or faded areas, we used the comparison function with the master image, and for character recognition (recording and correctness judgment), we utilized the OCR function. We set up 26 inspection frames, requiring an inspection cycle of 1.95 seconds. The recording method can be done in CSV and image formats.

In the automotive parts sector, particularly with mission parts manufacturers, we have previously conducted simple verifications of multiple samples for our valued customers. This time, we will measure the diameter of holes in metal components. 【Inspection Setup and Results】 By using the "Dimension and Angle Inspection" feature of EasyInspector, it was possible to measure the dimensions of two holes in just 0.5 seconds. We set it up to measure the distance from the center coordinates of each circle. By comparing the coordinates of the centers of each circle, we detected that they were within the specified value (29mm ± 0.4mm), resulting in a "Pass" judgment indicated by the blue enclosure in the image on the right. 【Software and Equipment Used】 Software Used: EasyInspector Field of View: Approximately 43 x 24mm Minimum Size of Inspection Target: 2mm Number of Inspection Points: 8 Camera Resolution: 1.3 Megapixels Lens Focal Length: 25mm Distance from Lens to Product: Approximately 530mm Lighting: Surface Emission Lighting Distance from Lighting to Inspection Item: Approximately 40mm The current 'EasyInspector2' MS (MeaSure) package allows for inspections of [Position and Width Measurement] and [Angle Measurement].

We received feedback from an automobile manufacturer regarding issues with the left and right wheel pins being incorrectly installed. This time, we are considering an inspection method to distinguish between "R" and "L" before the parts are installed in the automobile products. The orientation of the bolt markings is inconsistent, and we are exploring the possibility of checking them while they are aligned on the parts feeder. 【Inspection Setup and Results】 By using the "Comparison with Master Image" feature of EasyInspector, we were able to determine the left and right bolts. We attempted to identify them using the R and L markings, but detection was not successful due to reflections. Therefore, in this verification, we reported that by inspecting the orientation of the threaded part of the bolt, we could successfully distinguish between the R and L bolts in a well-aligned situation on the parts feeder in 0.36 seconds.

We received an inquiry about nut measurement from a manufacturer of mission parts such as truck components. I believe that traditional methods of "image analysis" and "image processing" will not be completely replaced by AI (deep learning). Depending on the customer's wishes and operational conditions, we will consider more cost-effective proposals through free simple verification and report back. 【Inspection Settings and Results】 By using EasyInspector's "Dimension Angle Inspection" feature, we were able to measure one dimension in 0.52 seconds. In the left image, the good product was detected within the reference value and was deemed "pass." In the right image, when a different nut (with a different height) was attached, it exceeded the height reference value, resulting in a "fail" judgment.

We received an inquiry from a company that manufactures hose parts for automobiles and motorcycles. It was decided to conduct a simple verification of the presence or absence of marks on the workpieces, as well as their sizes and dimensions, and they sent us images. In authoritative verification, it is common to receive images to report on whether a simple verification can be conducted, and afterwards, we will carry out more specific verifications such as consultations on operational methods. 【Inspection Settings and Results】 By using EasyInspector's "Presence or Absence of Specified Color Inspection" function, we were able to inspect the presence or absence and size of three marks. Dimension measurement was conducted using the "Dimension Angle Inspection" function. The left image shows the detection of marks. The right image is an enlarged view of the dimension detection screen.

Automotive parts and other resin-molded components may vary in the presence of clips, colors, or positions due to differences in specifications. However, in the case of similar products, accidents have occurred where similar but different items were shipped because they looked almost identical. 【Inspection Settings and Results】 By using EasyInspector's "Presence of Specified Color Inspection" feature, we were able to detect differences in the color and position of clips in nine locations and determine similar-looking items (different products) in 0.24 seconds. Our company offers verification and support from technical staff on a daily basis. If you have any issues, questions, or uncertainties during operation, please feel free to contact us at any time.

The interior of vehicles, such as cars, can be customized in terms of color and decoration according to the grade. We received an inquiry regarding color matching from an industrial machinery manufacturer with whom we have had previous dealings. They sent us sample images, and we conducted a simple verification. Our company website offers a free trial of our inspection software online. You can also experience our inspection software that uses AI (deep learning) firsthand. 【Inspection Settings and Results】 Out of 45 images, 15 (5 each) were used as training images. We inspected a total of 45 images, including the 15 training images and 30 untrained images. The result was that all 45 images were correctly detected. The breakdown of the images is 11 silver, 19 red, and 15 white. 【Software Used】 Software used: DeepSky

This is an inquiry from a manufacturer of automotive parts and accessories. I believe that clearly printing details such as model numbers and manufacturing dates on delivered products is a common practice across various industries. Our software includes basic functions for saving inspection results in CSV format or as images. It allows for the recording of which parts are associated with which product numbers, making it useful in various industries. 【Inspection Settings and Results】 By using the "OCRPro" feature of EasyInspector, we were able to read characters in 26 locations. The reading took 1.87 seconds. For characters, detection was possible using either the OCR function or the OCR Pro function, while for marks and logos, comparison inspection with master images was utilized. The images show differences in RGB and color tones in the "comparison with master image" settings. As for recording methods, results can be saved in CSV format or as images. Information like that in Table 2 will be recorded in CSV format.

Metal products such as die-castings often reflect light, making inspection difficult in the past. With our inspection software DeepSky, released in 2020, it is now possible to operate even reflective metal products like die-castings with simple settings. 【Inspection Settings and Results】 By using DeepSky's inspection features, we were able to determine dents, impressions, and chip blockages in just 0.35 seconds. The image on the left shows the hole that should be present in a good product, while the image on the right is taken with the same color (white) as the good product's "hole," but it is accurately recognized as defective. The number at the top of the frame indicates the AI's confidence level (number of recognition points). 【Software and Equipment Used】 Software Used: DeepSky Learning Version Field of View: Approximately 56 x 42 mm Minimum Size of Inspection Target: 20 mm Number of Inspection Points: 1 overall Camera Resolution: 1.3 million pixels Lens Focal Length: 12 mm Distance Between Lens and Product: Approximately 150 mm Lighting: Flashlight (Chip blockage: spotlighting the processing hole)

Our image inspection software is being considered in various fields and industries. The combination of AI (deep learning) and image inspection can be beneficial even in high-level solution areas. We recommend efficient inspections with DeepSky. [Inspection Settings and Results] The results of the verification conducted with the samples you provided showed that it was possible to detect "dents, impressions, chips, and large holes" and determine pass or fail. This time, we used a software called DeepSky, which utilizes AI (Deep Learning) for inspection. By training the software on the areas we want to detect, it can adjust its own setting parameters and recognize them. In actual operations, various patterns (sizes and features) of defects are expected, so it is necessary to increase the number of training images.

This time, we received a request from an automotive parts manufacturer for a free evaluation of the presence or absence of scratches on the metal parts of automotive components, based on the images provided. They sent photos of the valve (automotive bulb) and documentation of the defects. [Inspection Settings and Results] By using the "Scratch Detection" feature of EasyInspector, we were able to detect multiple scratches sensitively in less than one second. The threshold settings can be adjusted by the customer to determine how much of the scratch is considered acceptable. [Software Used] Software: EasyInspector310 Number of Inspection Points: 1 The current 'EasyInspector2' color package can be used for [scratch and defect detection].

Metal products often have parts that appear symmetrical. Many of our customers use our visual inspection software to prevent careless mistakes of parts that can be mistakenly implemented due to left-right errors from being passed on to the next process. 【Inspection Settings and Results】 By using the "Color Comparison Inspection / Presence of Specified Color Inspection" feature of EasyInspector, we were able to set the position and color for the correct orientation and make a judgment in less than one second. If you provide us with a photo of the defects you want to identify, it may also be possible to verify based solely on the photo. We look forward to hearing from you even if you cannot send sample products. 【Software Used】 Software Used: EasyInspector310 Number of Inspection Points: 1

Automotive parts and other resin-molded components can vary in availability and position due to differences in specifications. However, in the case of similar products, accidents have occurred where similar but incorrect items were shipped because they looked almost identical. We will inspect the presence of clips and rivets. 【Inspection Setup and Results】 By using DeepSky, it was possible to determine the presence of clips and rivets. While it is challenging to make judgments when the items are rotated 90 or 180 degrees, detection was possible by slightly shaking the items in the rotation direction within the visible range and training the system with the shaken images as examples. 【Software and Equipment Used】 Software Used: DeepSky Learning Version Field of View: Approximately 685 x 549 mm Minimum Size of Inspection Target: 2 mm Number of Inspection Points: 6 Camera Resolution: 1.3 Megapixels Lens Focal Length: 8 mm Distance from Lens to Product: Approximately 900 mm Lighting: Indoor Lighting

Incidents have occurred where similar resin-molded parts, such as automotive components, were shipped as different products due to their almost identical appearance. We conducted defect detection tests for dents and other issues based on inquiries from trading companies. 【Inspection Settings and Results】 Based on the samples we received, we were able to detect dents. The learning process was executed for approximately 2,400 steps, and the graph converged in about 16 minutes. The time may vary depending on the specifications of the PC used. The main specifications of the PC used for verification are: OS: Windows 10 Home 64bit, CPU: Intel(R) Core(TM) i7-9750H, RAM: 16GB, GPU: NVIDIA GeForce GTX1660 Ti. In this case, defects such as dents are labeled as "anomalies." The left image shows the annotation, and the right image shows the detection frame.

Automotive parts and other resin-molded components can vary in availability and position due to differences in specifications. However, in the case of similar products, accidents have occurred where similar but incorrect items were shipped because they looked almost identical. In previous image inspections, it was considered difficult to detect the presence or absence of black clips on black workpieces, as in the current inquiry. 【Inspection Settings and Results】 We conducted inspections using a software called DeepSky, which utilizes AI (Deep Learning). By training the software on the areas we wanted to detect, it adjusts its own setting parameters and learns to recognize them. The left image shows the detection of the installation of a black clip, while the right image correctly determines that the black clip is not installed. 【Software Used】 Software Used: DeepSky Learning Version Number of Inspection Points: 6

Automotive parts and other resin-molded interior painted products often have a glossy and elegant finish. We are considering our inspection software to stably produce higher quality products. 【Inspection Settings and Results】 We were able to detect black specks of about 0.2 mm within the black glossy paint. To capture fine defects, we proposed using multiple cameras with a method that slowly conveys the items for magnified imaging. In this evaluation, lighting was a significant factor. We conducted the inspection using software that employs AI (Deep Learning). Traditionally, image inspection has been difficult for glossy or mirror-finished products with conventional inspection software, but with innovative lighting and DeepSky, easy setup for inspection has become possible.

Automotive parts such as resin-molded components require higher quality interior parts as the luxury of the vehicle increases. This inquiry is about automating the inspection of small "dents, thread scratches, and chips" on interior painted parts. 【Inspection Setup and Results】 Using DeepSky's inspection capabilities, we were able to determine "dents, thread scratches, and chips" on resin-painted products. To achieve high precision in the inspection, we wanted to capture defective areas in larger images, so we evaluated them by magnifying the display. To shorten inspection time, we are proposing an inspection environment where multiple cameras inspect simultaneously on a conveyor. The left image is the "training image" for the defects, and the right image is the "detection frame."

The engines for motorcycles and automobiles produced by the machinery manufacturer, along with the parts implemented in them (this time it was the cover panel), require advanced technology. To emphasize the development of human resources and the improvement of technology, we recommend automating the simplified visual inspection. 【Inspection Settings and Results】 By using the "Dimension Angle Inspection" feature of EasyInspector, we were able to detect dimensional differences in two locations and determine visually similar items (different products) in 0.75 seconds. A free trial version of EasyInspector is available on our company's download page. You can also use images taken with your smartphone. Please give it a try. 【Software Used】 Software Used: EasyInspector710 Number of Inspection Points: 2 The current 'EasyInspector2' MS (MeaSure) package can inspect using [Position and Width Measurement] and [Angle Measurement].

The manufacturer of FA machine systems for automotive parts contacted us through their website regarding the inspection of scratches on the tapered part of a flare fitting. They sent a sample, and we provided a report based on a simple inspection. 【Inspection Settings and Results】 By using the "Scratch Detection" feature of EasyInspector, we were able to detect a scratch on one flare fitting. The scratches on the flare fitting were detectable using ring lighting. In the left image, only one scratch was detected, and it was judged as a failure in 0.34 seconds. Adjustments to the settings are necessary for differences in diameter and length. 【Software and Equipment Used】 Software Used: EasyInspector710 Field of View: Approximately 52 x 41mm Minimum Size of Inspection Target: 1mm Number of Inspection Points: 1 Camera Resolution: 1.3 Megapixels Lens Focal Length: 25mm Distance Between Lens and Product: Approximately 215mm Lighting: Ring Lighting Distance Between Lighting and Inspection Item: 50mm above Current inspections can be performed with the 'EasyInspector2' color package for scratch and defect detection.

We received sample images from an automotive parts manufacturer. We will report whether the dimensions are within tolerance through a free simple verification. In the inspection process, automation is being promoted to pursue stable inspection accuracy and cost reduction. 【Inspection Settings and Results】 By using EasyInspector's "Dimension Angle Inspection" function, we were able to confirm the tolerance of a dimension at one location (left settings screen). The inspection cycle time is 0.62 seconds. We are measuring the edges of the hole and flat surface in inspection frames 001 and 002 (right detection screen). The software can calculate the difference in the X coordinate values from 002 to 001, and it is possible to convert the dimensions to mm. In the free sample evaluation, we will first conduct a simple verification using the received samples or images and report the verification results. In the simple verification, we will evaluate whether the desired detection/judgment is possible with our in-house equipment. If detection or judgment is possible in the simple verification, we recommend conducting a test (feasibility verification) assuming actual operation, and evaluating processing time and judgment accuracy. If conducting feasibility verification, you can choose whether to continue with our company (for a fee) or to use our rental equipment for verification at your company.

We received a contact from a manufacturer considering a poka-yoke device for forgetting to paint rubber hoses. When there is identification paint (white, yellow, red, green) on the end face of a black rubber hose approximately Φ20, it will indicate OK; if there is no identification paint, it will indicate NG during the inspection of body parts. Our website offers a free trial of inspection software online. You can actually experience the inspection software that uses AI (deep learning). 【Inspection Settings and Results】 It is possible to determine that there is one blue mark on the inspection screen. In the case of red, the count for "red" will be one, and if there is one each of white and green, they will be counted separately. Since we have trained the system to recognize the absence of marks as "none," the above image is judged to have one "none." At this stage, we are simply counting how many of each there are, so if there is more than one "none," we will set the system to determine it as a failure. 【Software Used】 Software used: DeepSky

We have received a simple free evaluation to determine the types of tires, such as automotive parts. This time, we tried sending photos, but sending actual samples would provide the most reliable results. In that case, you will need to send several good products and the defective work you want to detect. Please contact us for details. 【Inspection Settings and Results】 The left image shows the annotation process (the task of enclosing the parts you want to teach). The right image displays the OK screen. The tire text is being evaluated by DeepSky. We also have the capability to record images and CSV files, making integration with surrounding systems easy. The number of detectable types that can be registered per product type ranges from 1 to 1000 (here, "types" refers to classification names used during annotation, such as "screw," "desiccant," "tomato," etc. If the number of detectable types exceeds 100, the detection rate may decrease.) 【Software Used】 Software used: DeepSky

A manufacturer of industrial equipment, with whom we have had a relationship for some time, has been actively inquiring about "DeepSky" since its release. Our company is accepting verification and support requests from our technical staff on a daily basis. If you have any issues, questions, or uncertainties during operation, please feel free to contact us at any time. 【Inspection Settings and Results】 Out of 42 images, 21 were used as training images. We inspected a total of 42 images, consisting of 21 training images and 21 untrained images. The result was that we were unable to detect only one untrained image, while all others were successfully detected. Images that were blurry or had faint prints were generally able to be judged. If we can stabilize the captured images further, we believe the accuracy could improve. 【Software Used】 Software used: DeepSky

We received multiple verification requests from a truck parts manufacturer. In the free sample evaluation, we first conduct a simple verification using the samples or images provided and report the verification results. If requested, we recommend conducting tests (feasibility verification) assuming actual operation, where we evaluate processing time and judgment accuracy. If feasibility verification is to be conducted, you can choose to continue with our company (for a fee) or use our loaned equipment for verification at your company. [Inspection Settings and Results] We verified whether we could detect burrs using DeepSky with the samples provided. As a result, we detected burrs in 54 out of 55 points, with one point not detected. All other burrs were successfully detected. With DeepSky, we can set the undetected images as additional training images, allowing us to gradually improve accuracy with each occurrence of undetectable defects after operation.

We received an inquiry from a customer who manufactures automotive parts and industrial machinery regarding the settings of DeepSky. They asked, "I thought the system was searching the entire screen based on the characteristics of the annotated area I set, but do you also remember any tendencies about where it tends to appear in the field of view?" 【Inspection Settings and Results】 We arrange squares of the same color and size and annotate only the upper part for training (see the image in the top left). In the image on the top right, we can find only the upper part. In the lower left image, we can also find only the upper part. However, in the right image, we could not find anything from the lower part. If the system can only find data that closely matches the teacher due to overfitting, the situation changes a bit. However, the position of the items we want to detect, such as defects, greatly affects inspection accuracy. It is necessary to use teacher images with various positions, orientations, and angles. As of 2021, DeepSky has been equipped with a teacher image augmentation feature. We can augment images of defects that rarely appear by flipping or rotating them in all directions and using brightness or Gaussian noise.

Press felt parts for automotive components may vary in the presence or absence of holes and the positioning of parts due to differences in specifications. However, in the case of similar products, accidents have occurred where similar but incorrect items were shipped because they looked almost identical. This time, we will verify whether the installation position of the clips is correct. 【Inspection Settings and Results】 We were able to distinguish between two types of clips in the overall view. It seemed challenging to determine the presence of the fastener's tacker or the text on the label from the overall view, so I suggested that using two cameras for one workpiece would be a more practical approach. The image shows the setup for inspecting whether the correct type of clip is in the correct position. The outer frame indicates the correctness of the installation position, while the inner frame determines whether the part is correct. 【Software Used】 Software Used: DeepSky

This is an inquiry from a manufacturer of resin-molded parts for automotive interior components, with whom we have had transactions in the past. Detecting black foreign objects or parts on a black background has been considered a challenging inspection task until now. The image inspection software DeepSky, which uses AI (deep learning), is capable of recognizing such similarly colored parts and foreign objects. 【Inspection Settings and Results】 Using a 1.3-megapixel camera, we captured approximately 1/4 of the overall field of view and conducted learning and detection. As a result, it was possible to identify the types of clips and the number of tackers under the verification environment. However, for items like fasteners attached to sloped surfaces where tackers are difficult to see, detection is likely to be challenging, so adjustments to the camera angle will be necessary. We categorized the labels into 11 types (naming and registering the items we want to find). We conducted simple tests after training with 13 teacher images for 1600 steps. 【Software Used】 Software used: DeepSky

The curtains installed inside vehicles such as cars, buses, and trucks are folded in an accordion style. If the various components, such as the belts called "hooks" and "tassels," as well as "Velcro," are not attached in the specified positions, the product will not pass inspection. There is a request to automate the current process where workers use jigs to inspect multiple times. 【Inspection Settings and Results】 The inspection was conducted using the inspection software "DeepSky." First, the components to be identified were annotated (left image). To help the software recognize the inspection areas, we annotated 10 good sample images divided into five categories: belts, part A, part B, Velcro, and product tags. Next, the areas where these components should be installed were specified (right image). Since the curtains are accordion-style, a proposal was made to use jigs for fixing in order to capture the workpiece in the correct specified position. It was possible to accurately determine whether the specified components were in the designated locations, marking it as OK if present and NG if not.

It has been said that determining the presence or absence of black components in black workpieces through image inspection is difficult. This time, we conducted an implementation inspection of black accordion-style curtains installed on windows of cars, trucks, and buses. In Hamamatsu City, Shizuoka Prefecture, where our company is located, there are many factories producing automotive parts, and our inspection software is utilized by various automotive parts manufacturers. [Inspection Settings and Results] We conducted the inspection using software that employs AI (Deep Learning). By training the software on the areas we want to detect, it adjusts its own setting parameters and becomes capable of recognition. This inspection software can easily inspect "black defects on black workpieces" and "metal (silver gloss) with metal components or defects (silver gloss)" with simple settings. We encourage you to try it out on our website's "DeepSky Learning Service." We verified it using the black curtains for automobiles that you brought. It was possible to detect the number of installation parts and determine the installation positions of the components. The blue detection frame outside the image indicates a pass for the specified area, while the light green and light blue detection frames inside indicate a pass for the types of components.

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

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.

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.