Virtual Prototyping of Optical Systems: Leveraging ImSym and Optical Scattering Data

Before starting mass production of a new optical system, or even small-scale production, it is essential to validate the system with an accurate prototype.

The Importance of BSDF Measurements in Virtual Prototyping

Integrating Bidirectional Scattering Distribution Function (BSDF) measurements in virtual prototypes helps engineers create reliable, high-quality optical products that meet technology demands. BSDF describes the complete scattering behavior as a function of source incident angle and wavelength:

In practice, this phenomenon is usually split into reflected and transmitted components, which are then treated separately as BRDF (Bidirectional Reflectance Distribution Function) and BTDF (Bidirectional Transmittance Distribution Function).

The following image shows the link between the roughness of a material and light reflection. By measuring the direction of the reflected light and obtaining the BSDF value, we can accurately model the roughness of a surface. 

End-to-End Virtual Prototyping with ImSym

The ImSym – Imaging System Simulator software product transforms the way optical engineers design and validate imaging systems by providing an end-to-end, physically accurate virtual prototype. ImSym models the entire imaging chain, including the optics, sensor, and image processing pipeline. This system-level approach provides an integrated view of the system's performance with simulations that translate into production-ready designs.

As illustrated in the following example, BSDF measurements can be incorporated into ImSym simulations to obtain precise visualizations of how light behaves when it interacts with surfaces and materials in the imaging system model.

Using ImSym with BSDF Data for Machine Vision Design

In modern factories, cameras are used with machine vision technology to inspect products and detect minute flaws and inconsistencies. Virtually prototyping machine vision optics can help evaluate factors such as determining the required camera speed to meet requirements, or troubleshooting unwanted stray light that is reaching the image sensor.

In this example, we want to simulate a machine vision camera that will be used to inspect a printed circuit board (PCB). The ImSym workflow to perform the machine vision camera simulation includes the following steps:

• Scene Object: Start with the PCB image file, corresponding to the ambient light, and then add a light source. This includes specifying the geometric and material properties of the PCB, which is crucial for accurate optical simulations. 
• Lens System: Model the lens system used in the optical setup. ImSym provides advanced tools for modeling complex lens assemblies, including aspheric lenses and freeform optics. You can choose lenses that can have varied properties such as focal length, aperture size, and lens materials. You can choose the right lens to optimize the optical performance and achieve the desired imaging results.
• Stray Light Analysis: Model the stray light properties of the optical system.
• ImSym provides the tools and analysis techniques necessary to evaluate the stray light properties of the optical system and illumination conditions. Stay light analyses using vary complexities from first order models to very detailed as-built opto-mechanical housings can be conducted.
• Detector: Define the properties of the detectors used in the optical system. This includes modeling the type of detector (e.g., CCD, CMOS), its resolution, sensitivity, and other relevant characteristics. Accurate detector modeling is essential for simulating how the optical system captures and processes light.
• Image Signal Processing: Simulate the image signal processing (ISP) pipeline. This includes steps such as noise reduction, color correction, and image enhancement. By modeling the signal processing, you can predict how the final image will be affected by various processing algorithms and optimize them for better image quality.
• Simulation (Final Image): Display the final image generated by the optical system, as modeled by ImSym. This provides a visual representation of the system's performance, allowing you to evaluate the quality of the image and identify any potential issues. You can analyze the image and make adjustments to the design to achieve the desired system performance.

The following figure shows the initial ImSym simulation of the machine vision camera. Some holes in the PCB are circled in red, which indicates that they were successfully detected. Those that were not successfully detected remain without red circles. The specular reflection of the copper on the PCB leads to saturated pixels in the camera, preventing the detection of the holes. This oversaturation occurs because the highly reflective surface of the copper creates intense glare, which overwhelms the camera sensor. Consequently, critical features such as holes are obscured and cannot be accurately identified.

One way to address this issue is to include an optical scattering measurement of the green PCB and copper in the ImSym simulation, which provides a more realistic view of the interaction between light and the PCB. The simulation can account for both specular and diffuse reflections, providing a comprehensive representation of how light behaves when it encounters the copper surface. This will give you additional insights so that you can also adjust the detector and ISP model used in ImSym to improve detection of PCB details.

You can use the Synopsys Mini-Diff V2 instrument to obtain the BSDF data.

The copper reflection is not purely specular; there is also a diffusion around the peak. This indicates that the reflective properties of the copper surface are more complex than initially anticipated.

The specular reflection, which is the mirror-like reflection of light, is accompanied by a diffuse component that scatters light in various directions. This scattering can affect the accuracy of detecting features, such as holes on the PCB, since the diffused light may obscure or blur the details that need to be identified.

Adding the BSDF data measured by the Mini-Diff V2 to the ImSym model and adjusting the detector and ISP parameters helps to mitigate issues related to oversaturation and enhances the ability of the optical system to detect fine details on the PCB. As a result, the inspection process is significantly more accurate and reliable, leading to better quality control and more reliable PCB manufacturing. The updated ImSym simulation is shown below, in which all the PCB holes are detected.

Understanding and accounting for both the specular and diffuse components of the reflection are crucial for improving the detection algorithms and ensuring precise inspection of the PCB.

Accelerating Imaging Design Innovation

ImSym allows engineers to model virtual prototypes of optical systems, providing early warnings of issues and enabling direct testing of image quality and signal processing. This proactive approach saves time and reduces costs by identifying problems early.

ImSym models the entire image acquisition chain and can use component models of varying fidelity. Synopsys optical scattering instruments measure BSDF, precisely describing light interactions with materials. Integrating measured data into ImSym creates an increased level of realism for reliable virtual prototypes, or virtual twins, closely mirroring the physical prototype.