By Dr. Miguel El Guendy, Technology Analyst at IDTechEx
The use of image sensors is set to revolutionize multiple industries from advanced driver-assistance system (ADAS)/autonomous driving to agriculture, industrial inspections, and medical imaging. The visualization of images beyond the visible light spectrum and the traditional red, green, and blue (RGB) is gaining traction and is set to impact the future of sensors in many more applications.
The new IDTechEx report, “Emerging Image Sensor Technologies 2024-2034: Applications and Markets“, explores a diverse range of image sensing technologies capable of resolutions and wavelength detection far beyond what is currently attainable. IDTechEx expects that the growing importance of autonomous technologies will lead the emerging image sensor market to over US$739 million by 2034.
There are several unique selling points to adopting these types of image sensors, one of which is the broadening of the spectral coverage over the commonly used CMOS RGB sensors.
CMOS detectors are highly effective and reliable for visible light image capture and are widely used in robotics, industrial inspections, and consumer electronics applications. Their cost-effectiveness and ease of implementation into a variety of applications ensure successful adoption in the market. Nevertheless, with a growing need to visualize a wider range of frequencies in the electromagnetic spectrum comes a new market opportunity with immense potential.
One of the areas of interest is capturing images in the short-wave infrared (SWIR) region (1000-2000nm). The autonomous vehicle industry, for instance, is particularly invested in technologies operating in this region, as visible light image capture is less effective in foggy/misty/dusty/dark conditions. Similarly, industrial inspections could benefit from SWIR imaging as this region has different absorption ratios that can help distinguish materials that look the same in the visible spectrum.
The main challenge, however, is lowering the costs associated with the existing technologies in this space. The dominant technology in this sector is based on indium gallium arsenide (InGaAs), and while these devices perform well and have high sensitivity in these frequency ranges, they are also costly, with average pricing approximating US$20,000. As a result, they are used on very specialized applications and remain inaccessible for autonomous driving and consumer electronics, for instance.
The potential demand for more affordable SWIR image sensors has triggered the emergence of novel technologies that are determined to fill this customer need. These include hybrid image sensors, where organic semiconductors or quantum dots are placed above a CMOS readout integrated circuit (ROIC), and extended silicon, where the physical properties of silicon are altered, enabling its absorption range beyond its bandgap capabilities. Despite promises of performance enhancement, these technologies require further maturity to compete with InGaAs.
Nevertheless, there is more innovation happening in the image-sensing space. Companies are not solely looking at expanding the spectral range; there is also ongoing work in imaging over a larger area whilst increasing temporal resolution and dynamic range. An example of this is event-based vision, where each pixel in the sensor detects changes in intensity and sets a timestamp reporting this change. Rather than using a frame-based approach that would require extensive computational processing, this method increases the temporal resolution for a fraction of the computation.
The growth of the Internet of Things (IoT) and smart consumer electronics is driving another phenomenon in the image sensor sector, which is the miniaturization of these sensors to conform within the smallest form factor devices. Miniaturized spectrometers, for instance, target these applications and aim to deliver sufficiently small and low-cost solutions to the market.
There is one main trend that remains consistent irrespective of the frequency sensitivity ranges, and this is cost. While CMOS sensors are already relatively mature and affordable, there is ample opportunity to deliver similar products for other frequencies in the electromagnetic spectrum. The challenge consists in lowering the cost of manufacturing and the form factor whilst replicating or even surpassing the performance of CMOS in the visible light region for other frequency ranges in the electromagnetic spectrum. IoT devices are getting increasingly smaller, and thus, sensors fitted into these devices should conform to their shape and draw minimal power, which is also an added difficulty, considering many high-resolution image sensors have high computational requirements associated with them.
The world of image sensors has enormous potential and could impact multiple industries, from autonomous vehicles to industrial inspections and agriculture. The fulfillment of its potential is dependent, however, on its cost and performance, and while there are currently many technological players in the running for dominance in the sector, the technology that delivers on these points most suitably will earn a significant share of this large market, that in the future will supply to autonomous vehicle manufacturers and IoT providers in the consumer electronics space.
