Spectroscopic Imaging for Smartphone

Technology Overview

With our technology, spectral data are obtained as spatial fringe patterns. As the interference phenomena are able to occur in narrow gaps, the extremely narrow space inside of this device was sufficient. As a result, we are able to realize an extremely small hyperspectral camera.

We had developed two types of prototypes of our proposed spectroscopic imager. One is two-dimensional spectroscopic imager. Another one is one-dimensional type, known as one-shot type, which is able to obtain the spectroscopic character distribution on a line with a single frame data.

The one-shot type can be used to realize a compact line spectral imager in the visible, near-infrared and mid-infrared lights region on a smartphone. 

Technology Features & Specifications

The technology comprises a video camera, software algorithm and the Hyperspectral camera unit with basic analysis software. Its specifications are

  • Operating wavelength: visible light(0.4µm-0.9µm), near-infrared light(0.9µm-2µm), mid-infrared lights(8µm-14µm)
  • Size: Palm-size for the two-dimensional model (high sensitivity) and finger-size for the one-shot model (high resolution)
  • Able to obtain 60 spectra in 1 sec using a camera running at 60 fps
  • Accuracy can be improved using temporal averaging

For evaluation of the accuracy, we measured the extremely low glucose concentrations that corresponded to human blood-sugar level. We used 4 kinds of glucose water solution (concentration range: 50 mg/dl ?200 mg/dl). The super continuum (Maker: Fianium?Type: WhiteLase SC480-2?Wavelength range: 480-2400 nm) was used as broadband near-infrared light-source. Other experimental conditions were InGaAs camera (Maker: Hamamatsu Photonics K.K., Type: C10633-13, Detective wavelength region: 900 nm-1700 nm). We could confirm the high correlation between absorbance at 1600 nm and glucose concentration. Then, we calculated the temporal average of 60 measurement data within 1 sec. From this evaluation results, we could reduce 3σ into 1/7 and discriminate each glucose concentration with correlation coefficient 0.92. Therefore, we verified the temporal averaging effect with high time resolution measurement and demonstrated the feasibility of high measurement accuracy.

Potential Applications

  • Healthcare sensor
  • Medical sensor
  • Environmental sensor

Market Trends and Opportunities

The global hyperspectral imaging (HSI) market is expected to reach $ 74.3 million by 2019, projected to grow at a CAGR of 12.0% during the forecast period of 2014 to 2019. HSI technologies are widely used in areas such as mining & mineralogy, agriculture, life science & diagnostics, and environment testing.

Customer Benefits

  • The One-shot model comes at an affordable cost
  • Compact in size and is therefore suitable for portable applications

Make an Enquiry