Chip HardwareWavelength range, number of spectral channels, spatial resolution, package type, data interface, etc.

Software AlgorithmsSpectral acquisition algorithm, reconstruction algorithm, automatic white balance, image quality optimization, etc.

As imaging technology moves into an era of refinement, accurate color reproduction and stable performance across diverse scenarios have become core competitive strengths of smart hardware imaging capabilities. Focusing deeply on multispectral imaging, Seetrum delivers a full-chain multispectral imaging solution based on self-developed computational spectral technology. The solution covers core sensors, module R&D, algorithm models and calibration systems. Boasting high performance, low power consumption and great compatibility, it brings outstanding color performance to smart imaging devices such as smartphones and action cameras.

I. Multispectral Chips

Seetrum’s consumer-grade multispectral imaging chips cover the core wavelength range of 380~960 nm with up to 12 physical spectral channels. Programmable functions include frame rate adjustment, image flipping, windowing and multiple binning modes, supporting standard data output such as 10-bit RGB RAW.

The chip integrates OTP memory, static bad pixel correction and automatic black level calibration. Equipped with dual PLLs and multi-camera synchronization interfaces, it ensures excellent imaging consistency and multi-device collaboration. Adopting the advanced Stacked BSI pixel architecture, the chip runs steadily at 60 FPS across resolutions from 0.5 MP to 1.3 MP, with a minimum power consumption of only 15 mW. It balances high performance and low power consumption, perfectly matching the slim profile and long battery life requirements of smartphones, action cameras and similar devices.

II. Spectral Algorithms & Applications

In terms of core algorithms and applications, Seetrum has built a full-process solution spanning spectral data acquisition, spectral reconstruction, automatic white balance and image quality optimization.

Our self-developed global multispectral automatic white balance algorithm accurately derives key parameters including light source spectrum, light type and correlated color temperature, supporting both color constancy and chromatic adaptation output modes. The one-click reconstruction and rendering tool enables intuitive and efficient imaging effect evaluation.

Verified in full-scenario tests including indoor, outdoor, multi-light source and solid-color background environments, our solution significantly reduces inherent color deviation of original imaging and delivers results highly consistent with ideal reference standards. It also effectively addresses the industry challenge of color inconsistency among multiple cameras.

III. Spectral Calibration System

To resolve industry pain points regarding manufacturing processes and consistency of multispectral sensors, Seetrum has developed an innovative calibration large model for enhanced spectral consistency, breaking through technical bottlenecks in high-dimensional spectral data representation and high-performance implementation with mainstream production processes.

Our self-developed calibration algorithms feature strong generalization capability. In actual tests covering 1,092 modules, the optimization rate of spectral and correlated color temperature calibration performance exceeds 83% and reaches up to 90%, driving improvements in both product yield and production efficiency.

Furthermore, we have established a full-process calibration system combining local and cloud platforms that covers all stages of module production. The standardized and intelligent calibration solution guarantees uniform performance of every single device.