Types of machine vision lighting control
Most machine vision systems typically use one of three types of light source drivers: constant voltage drive, pulse width modulation (PWM) voltage drive, or constant current drive. The earliest light source controllers were voltage-driven, a technique sufficient for most applications. However, voltage-driven light sources have significant drawbacks: the power supply voltage is constant, and the brightness is determined by the current, not the voltage. Furthermore, because the light source's temperature changes during use, the same voltage will produce different brightness levels.
The figure illustrates the linear relationship between illuminance and forward current for a typical light source. It shows that the relationship between forward current and voltage is highly non-linear; a small change in the supply voltage can cause a large change in current, resulting in a large change in brightness. This effect means that voltage-driven systems cannot achieve precise and accurate regulation of lighting intensity. Voltage-driven systems are also susceptible to small fluctuations in the system supply voltage, which can lead to significant changes in lighting levels.
Despite the limitations of voltage-driven lighting, some lighting controllers utilize this type of power supply to save costs. Since voltage-driven lighting cannot directly control brightness, pulse-width modulation (PWM) can be used instead. PWM switches on and off several times per exposure to achieve an average brightness. The PWM pulses are controlled by an internal clock, and a faster internal clock is required for precise brightness control. The clock speed limits the number of lighting levels that PWM can provide. A typical voltage-driven lighting controller offers 100 intensity levels, while a typical current-driven lighting controller typically offers thousands of intensity levels and a much better level of control.
The design of a constant current lighting controller directly controls the lighting intensity through a controllable current. Constant current driving is perhaps the most precise and reliable choice for LED lighting control because lighting intensity is proportional to current, and current driving is unaffected by power supply variations. Constant current controllers also offer significant advantages in high-speed systems and applications requiring pulsed light.
Machine vision lighting control principle
Machine vision lighting control principles mainly involve the precise adjustment of parameters such as brightness, color, and direction of the light source to meet the image quality and sharpness requirements of different application scenarios. The following are some key lighting control principles:
Brightness control: The brightness of a light source can be controlled by adjusting its power or current. The goal of brightness control is to ensure that the image has sufficient contrast so that the machine vision system can accurately identify and measure the target object.
Color control: Machine vision systems may require light sources of specific colors to highlight certain features of a target object or reduce background interference. Therefore, color control can be achieved by selecting appropriate light source types and color filters.
Directional control: The direction of light from the source has a significant impact on image quality. By adjusting the position and angle of the light source, the reflection and projection of light onto the target object can be controlled, thereby optimizing the visual effect of the image.
In practical applications, machine vision lighting control systems typically adjust the illumination conditions of the light source automatically based on preset parameters or real-time image feedback. For example, when the color or surface characteristics of the target object change, the system can automatically adjust the color and brightness of the light source to ensure the stability of image quality.
There are also advanced lighting control technologies, such as machine learning-based adaptive lighting control. This technology can optimize lighting conditions based on large amounts of image data and machine learning algorithms, further improving the performance and accuracy of machine vision systems.
What are the illumination methods for machine vision light sources?
There are various illumination methods for machine vision light sources, each with its specific application scenarios and advantages. Below are some common machine vision light source illumination methods:
Ring or spot lighting: This method is primarily used for immediate illumination, where the light source shines directly onto the object to obtain a clear image. This lighting method is very effective when a negative color image of the object is required. However, it can cause specular reflections when the light shines on glossy or reflective materials.
Coaxial illumination: Coaxial light is generated by the change and convergence of light emitted from a vertical wall. It is then projected onto a beam splitter that directs the light downwards, and a camera observes the object from above using a beam splitter. This method is mainly used for surface imaging inspection of semiconductors, PCBs, and metal parts, as well as for measuring the shape and size of tiny components.
Dark-field lighting: This method provides low-angle illumination relative to the surface of an object, primarily used to highlight certain features or variations in surface texture. If the light source is not visible in the line of sight, it is considered dark-field lighting; the position of the light source determines whether it is bright-field or dark-field lighting.
Diffuse illumination: Continuous diffuse illumination is primarily used when the transmittance of a target slope is high or when the slope has complex viewing angles. This illumination method is very effective for illuminating fully mounted circuit boards, providing uniform illumination within a 170° solid angle area.
In addition, there are other illumination methods, such as vertical lighting, backlighting, multi-angle lighting, and bowl lighting. Each method has its specific application; for example, vertical lighting is suitable for uniform illumination of a large area, while backlighting can provide uniform field of view for observing the side profile of an object.
In general, there are many ways to illuminate machine vision light sources, and the choice of which method to use depends mainly on the specific application requirements, the characteristics of the object, and the required image quality.
