Search

Long-range (~250m) measurement systems with 3D resolution are highly anticipated for various applications such as automotive, surveillance and robotics systems. Direct time-of-flight (ToF) systems based on CMOS image sensors (CIS) with avalanche photodiodes with the Geiger-mode or single-photon avalanche diodes (SPAD) have been developed for these purposes. One difficulty is to meet practical requirements of long-range (~250m) measurement capability together with both lateral and depth resolution. Although indirect-ToF systems based on modulation and phase sensitive detection have fine depth resolution of the order of sub-cm, their full range is limited only to ~4m with standard Si-photodiodes [1] and ~20m with the gain assist of APD [2]. Direct-ToF systems based on SPAD pixels with time-to-digital converters (TDC) have long range measurement capability but have not reached megapixel resolution [3], [4]. Another direct-ToF of sub-range syntheses (SRS) system, which detects single photons returned from an object located in each sub-range by synchronous gating, was demonstrated to have a long range (250m) measurement capability with high lateral resolution of 10cm. However, it has a depth resolution limited by the width of optical pulse source [5]. In order to circumvent this issue, in this work, we develop an SRS-ToF system based on a 6µm pitch, 1200×900pixels, Geiger-mode operated vertical avalanche photodiodes (VAPD) CIS that enables one to configure depth resolution down to 10cm for short-distance (

We have developed a direct time-of-flight (TOF) 250 m ranging Complementary Metal Oxide Semiconductor (CMOS) image sensor (CIS) based on a 688 &times, 384 pixels array of vertical avalanche photodiodes (VAPD). Each pixel of the CIS comprises VAPD with a standard four transistor pixel circuit equipped with an analogue capacitor to accumulate or count avalanche pulses. High power near infrared (NIR) short (&lt, 50 ns) and repetitive (6 kHz) laser pulses are illuminated through a diffusing optics. By globally gating the VAPD, each pulse is counted in the in-pixel counter enabling extraction of sub-photon level signal. Depth map imaging with a 10 cm lateral resolution is realized from 1 m to 250 m range by synthesizing subranges images of photon counts. Advantages and limitation of an in-pixel circuit are described. The developed CIS is expected to supersede insufficient resolution of the conventional light detection and ranging (LiDAR) systems and the short range of indirect CIS TOF.

We have developed a direct time-of-flight (TOF) 220 m ranging CMOS image sensor (CIS) based on a 688 × 384 pixel array of vertical avalanche photodiodes. An area scene at 220 m ahead can be ranged and imaged by one high power light pulse illuminated through diffusing (without scanning) optics and sub-photon level signals are extracted by in-pixel pulse count circuits. Depth map imaging with a 10 cm lateral resolution and with a 30 cm depth precision is realized from 1 m to 220 m range. The developed direct-TOF CIS with SPSE is expected to supersede the insufficient resolution of the conventional light detection and ranging (LiDAR) systems and the short range of indirect CIS TOF.

We developed multiocular 1/3-inch 2.75-μm-pixel-size 2.1M- pixel image sensors by co-design of both on-chip beam-splitter and 100-nm-width 800-nm-depth patterned inner meta-micro-lens for single-main-lens stereo camera systems. A camera with the multiocular image sensor can capture horizontally one-dimensional light filed by both the on-chip beam-splitter horizontally dividing ray according to incident angle, and the inner meta-micro-lens collecting the divided ray into pixel with small optical loss. Cross-talks between adjacent light field images of a fabricated binocular image sensor and of a quad-ocular image sensor are as low as 6% and 7% respectively. With the selection of two images from one-dimensional light filed images, a selective baseline for stereo vision is realized to view close objects with single-main-lens. In addition, by adding multiple light field images with different ratios, baseline distance can be tuned within an aperture of a main lens. We suggest the electrically selective or tunable baseline stereo vision to reduce 3D fatigue of viewers.

We present a CMOS image sensor that enables a compact 3-dimensional (3D) vision camera system comprising a single set of the sensor and a camera lens. In order to make binocular parallax, which is essential for 3D imaging, the input pupil of the camera lens is presumed to consist of the right-eye and the left-eye domains, where the pixels exclusively receiving light beams from the right-eye domain and those from the left-eye domain, are arranged alternately. In addition, the sensor features an on-chip lenticular lens to split the incident light from the two directions and a Digital Micro Lens [1,2] to focus the split light beams onto the dedicated pixels without significant crosstalk. The fabricated 3D image sensor enables not only successful stereovision imaging in color with sufficiently high sensitivity, but also accurate calculation of distance.

We present a day and night MOS imager based on a single plate on-chip interference color filter. The filter comprises periodic multiple layers of TiO 2 and SiO 2 , with an intermediate color selection layer (SiO 2 ) to disturb the period of the layers, analogous to a "defect" layer in the one-dimensional photonic crystal. A particular advantage of this filter is flexibility of designing a spectral profile of each color. Thus, one unit cell of the present MOS imager is designed to have three multi-spectral, i.e. R+IR, G+IR, B+IR, pixels and one IR dedicated pixel, which would never be realized by ordinary pigment materials. Daytime color image signals are obtained by subtracting the IR pixel signal, as a reference, from each signal of R+IR, G+IR and B+IR pixels. Nighttime black and white imaging is simply realized by using the IR components of all the pixels as brightness signals. This enables seamless switching between the day and night operations of a camera. Although the subtraction operation usually reduces the dynamic range (DR) and signal-to-ratio (SNR), in particular at low color temperatures, we overcome the issues by employing a new design scheme of the color filter comprising double defect layers for each visible pass band and narrow IR pass bands for common IR components. As a result, signal degradations in SNR and DR are suppressed even at low color temperatures enabling daytime imaging in a wide range of color temperatures from 2300 K to 6500 K.