Industrial Cameras Case Studies of Products in Use
Image Sensor Positioning Solutions
- Industrial Cameras
- Factory Automation Cameras
- Machine Vision Cameras
- Automatic adjustment
- High precision
- Product miniaturization
- Efficiency enhancement
- High-mix low-volume production
- Production efficiency
- Employee retention
- Work qualit
- Image sensors
Image processing enables high-precision positioning and automatic adjustment
At manufacturing sites, electronic parts are constantly being made smaller and with greater precision at an ever faster pace. As a result, there is a demand for higher performance in the functionality required for industrial cameras, and this need is rapidly increasing.
Also, rather than mass producing the same product, the diversification of products is shifting the paradigm to high-mix low-volume production.
Manufacturers that do high-mix low-volume production have wanted to increase work efficiency along with improving quality.
In addition, in recent years the cost of labor has been increasing and birthrates are declining as the population ages, which means that employee retention is becoming a big issue.
We solved these problems by using image processing to enable image sensors to position parts automatically.
Automatic positioning solutions by using image sensors through industrial cameras can achieve three things: Higher precision industrial cameras; Reduction of personnel by automating adjustments; and Product size reduction.
Manual Adjustment of Image Sensor
The production facilities used by manufacturers need to be highly accurate at all times. Industrial cameras, which are used to replace human observers, are essential to automated equipment in particular, and also require very advanced assembly technology.
Industrial cameras, which are used to align the positions of target objects, must be designed so that the product in the center of the lens and the center of the image sensor align accurately. To align the optical axis of the camera with the reference axis of the image sensor during assembly, a measurement chart was shot by the subject image sensor, and the worker adjusted the position while confirming the image, and when the final stage was complete, the screws were tightened.
We could not work efficiently and used a lot of time on adjustment work. This was due to the position deviating when the screws were manually tightened, the complexity of the work procedure, and the amount of experience of the worker who was manually adjusting the position.
Automatic Positioning Using Image Processing
The positioning work was done automatically by capturing images from the subject camera frame and image sensor, then using image processing to calculate the amount of deviation, and then feeding back those results to the mechanical components.
By automating the screw tightening process, we corrected the position deviation caused by tightening the screws, so we could get more precise positioning.
The worker just sets up the parts and presses the start button to complete the highly precise positioning work, unaffected by the skill of the person in charge of the work.
By using the image, we were able to achieve automatic positioning at the micron order, which people cannot do by hand. In addition, we also succeeded in improving work efficiency by achieving many other results, such as reducing manpower by automating adjustments, product size reduction, and reducing the number of parts.
We selected the optimal cameras and lenses to ensure image sensor positioning accuracy of 25 μm.
To do this, we used a small and lightweight USB camera, a microscopic objective lens to get sub-micron level optical resolution, and lighting to more clearly capture the characteristics of the subject. This system is also capable of high-speed transmission of large image files.
We achieved positioning at ±25 μm precision by using the following equipment configuration.
- Camera: BU505MG (manufactured by Toshiba Teli)
Pixel count: 2448 (H) × 2048 (V)
Pixel size: 3.45 μm × 3.45 μm
- Lens: Objective lens for bright field
- Lighting: Coaxial epi-illumination (fiber lighting)
- XY positioning accuracy: ±25 μm or less
- θ accuracy: ±0.07° or less
- Flange back accuracy: 17.526 mm ±50 μm or less
- Tact time 0.1 H or less (6 min or less)
DU Series, DDU Series, CSC6M100 Series, CSCS60BM18
Sharpness in the specified area is calculated and used as the focus value. The focus value is calculated as the Z stage moves, and the stage stops where the focus value reaches its peak.
- Edge Detection
Detects the edge that is the reference for the camera frame.
- Pattern Matching
Registers the pattern in the image sensor needed to decide the position, and then searches for the specified pattern through the targeted image sensor.
- Anti-vibration table
The slightest vibration affects the micron-order positioning.
This equipment has an anti-vibration table to enable high-accuracy positioning work to be done automatically.