Let me briefly talk about the principle. Taking the agricultural and sideline rice as an example, as shown in Figure 1, rice enters from the feed inlet, enters the detection area through the chute, the camera performs real-time image collection, and after image recognition, differentiate normal rice, heterochromatic rice and impurities by color, and control the air nozzle in real time to separate heterochromatic rice and impurities from normal rice.In general, the color sorter is a modern, high-tech food processing equipment that integrates light, mechanical, electrical, pneumatic and other technologies.
Let's clarify the main technical indicators of the color sorter
(1) Processing speed
That is, the number of materials that can be processed per unit time. The main factors affecting this index are the speed of the servo system, the image processing capability of the visual system, and the maximum speed of the conveyor belt.
(2) Color selection accuracy
Color sorting accuracy refers to the proportion of features separated from the mixture to total features, reflecting the accuracy of the color sorter. This parameter is mainly affected by the recognition accuracy of the image system, the accuracy of the servo system when separating raw materials, and the conveyor Influenced by running speed.
(3) Take out ratio
The take-out ratio refers to the ratio of the number of features and non-features selected by the color sorter. The take-out ratio can be adjusted. The adjustment of the take-out ratio mainly depends on the adjustment of the energizing time of the actuator. The influencing factor is mainly the mechanical accuracy.
Let ’s talk about the processing speed first. Looking at the visual part alone, to get a higher sorting speed, you need the lens to have a sufficiently large aperture. You can obtain a larger frame rate by reducing the camera exposure time. There must be enough light to get an image that meets the illumination in a short exposure time; secondly, according to the optical properties of the object to be measured, a reasonable distribution of light source brightness and color can also improve the sorting efficiency.
Finally, the accuracy of color selection is mainly reflected in the ability of the lens and camera in color recognition. Obviously, for two objects with very small color differences, the chromatic aberration of the lens must be small and the sharpness must be higher. This type of lens is mainly used for screening large-size objects, so the resolution requirements are relatively low, and only need to have a high contrast at low frequencies. The second is the camera. Using a color camera requires a camera with high color reproduction, a wide response spectrum, and a large pixel size.
To meet the special needs of the color sorting industry, DZO has developed LM series line scan lenses. After testing, the use of LM5022A (50mm focal length, F2.2 aperture) combined with 2K / 4K / 8K line scan camera for color selection operation can achieve excellent detection results. This lens has excellent color recognition capabilities and high image restoration The lens aperture is large, which is very suitable for high-speed shooting.