We have been researching and developing a CMOS image sensor that has 2.8 μm x 2.8 μm pixel, 33-Mpixel resolution
(7680 horizontal pixels x 4320 vertical pixels), 120-fps frame rate, and 12-bit analog-to-digital converter for “8K Super
Hi-Vision.” In order to improve its sensitivity, we used a 0.11-μm nanofabricated process and attempted to increase the
conversion gain from an electron charge to a voltage in the pixel. The prototyped image sensor shows a sensitivity of 2.4
V/lx•s, which is 1.6 times higher than that of a conventional image sensor. This image sensor also realized the input-referred
random noise as low as 2.1 e<sup>-</sup><sub>rms.</sub>
We have developed a CMOS image sensor with 33 million pixels and 120 frames per second (fps) for Super Hi-Vision (SHV:8K version of UHDTV). There is a way to reduce the fixed pattern noise (FPN) caused in CMOS image sensors by using digital correlated double sampling (digital CDS), but digital CDS methods need high-speed analog-to-digital conversion and are not applicable to conventional UHDTV image sensors due to their speed limit. Our image sensor, on the other hand, has a very fast analog-to-digital converter (ADC) using “two-stage cyclic ADC” architecture that is capable of being driven at 120-fps, which is double the normal frame rate for TV. In this experiment, we performed experimental digital CDS using the high-frame rate UHDTV image sensor. By reading the same row twice at 120-fps and subtracting dark pixel signals from accumulated pixel signals, we obtained a 60-fps equivalent video signal with digital noise reduction. The results showed that the VFPN was effectively reduced from 24.25 e<sup>-</sup><sub>rms</sub> to 0.43 e<sup>-</sup><sub>rms</sub>.
We have developed a back-side-illuminated image sensor with a burst capturing speed of 5.2 Tpixels per second. Its
sensitivity was 252 V/lux·s (12.7 times that of a front-side-illuminated image sensor) in an evaluation. Sensitivity of a
camera system was 2,000 lux F90. The increased sensitivity resulted from optical and time aperture ratios of 100% and
also by increasing from a higher optical utilization ratio. The ultrahigh-speed shooting resulted from the use of in-situ
storage image sensor. Reducing the wiring resistance and dividing the image area into eight blocks increased the
maximum frame rate to 16.7 million frames per second. The total pixel count was 760 horizontally and 411 vertically.
The product of the pixel count and maximum frame rate is often used as a figure of merit for high-speed imaging devices,
and in this case, 312,360 multiplied by 16.7 million yields 5.2 Tpixels per second. The burst capturing speed is thus 5.2
Tpixels per second, which is the highest speed achieved in high-speed imaging devices to date.
We have developed an ultrahigh-speed CCD camera that can capture instantaneous phenomena not visible to the human
eye and impossible to capture with a regular video camera. The ultrahigh-speed CCD was specially constructed so that
the CCD memory between the photodiode and the vertical transfer path of each pixel can store 144 frames each. For
every one-frame shot, the electric charges generated from the photodiodes are transferred in one step to the memory of
all the parallel pixels, making ultrahigh-speed shooting possible. Earlier, we experimentally manufactured a 1M-fps
ultrahigh-speed camera and tested it for broadcasting applications. Through those tests, we learned that there are cases
that require shooting speeds (frame rate) of more than 1M fps; hence we aimed to develop a new ultrahigh-speed camera
that will enable much faster shooting speeds than what is currently possible. Since shooting at speeds of more than
200,000 fps results in decreased image quality and abrupt heating of the image sensor and drive circuit board, faster
speeds cannot be achieved merely by increasing the drive frequency. We therefore had to improve the image sensor
wiring layout and the driving method to develop a new 2M-fps, 300k-pixel ultrahigh-speed single-chip color camera for
We developed a 300,000-pixel ultrahigh-speed CCD with a maximum frame rate of 2,000,000 frames per second. The
shooting speed of the CCD was possible by directly connecting CCD memories, which record video images, to the
photodiodes of individual pixels. The simultaneous parallel recording operation of all pixels results in the ultimate frame
rate. We analyzed a voltage wave pattern in the equivalent circuit model of the ultrahigh-speed CCD by using a SPICE
simulator to estimate the maximum frame rate. The pixel area was consisted of 410 and 720 pixels in the vertical and
horizontal and divided into 8 blocks for parallel driving. An equivalent circuit of one block was constructed from an RC
circuit with 410 × 90 pixels. The voltage wave pattern at the final stage of an equivalent circuit was calculated when a
square wave pulse was input. Results showed that the square wave pulse became blunt when the driving speed was
increased. After estimation, we designed the layout of the new ultrahigh-speed CCD V6 and fabricated the device.
Results of an image capturing experiment indicated a saturation signal level of 100% that was maintained up to 300,000
frames per second. A saturation signal level of 50% was observed in 1,000,000 frames per second and of 13% in
2,000,000 frames per second. We showed that the maximum frame rate is dependent on a drop of the saturation signal
level resulting from the driving voltage wave pattern becoming blunt.
A structure for backside illuminated ultrahigh-speed charge coupled devices (CCDs) designed to improve the light
sensitivity was investigated. The structure's shooting speed of 1 million frames/second was made possible by directly
connecting CCD memories, which record video images, to the photodiodes of individual pixels. The simultaneous
parallel recording operation of all pixels results in the highest possible frame rate. Because back-side illumination
enables a fill factor of 100% and a quantum efficiency of 60%, sensitivity ten or more times that of front-side
illumination can be achieved. Applying backside illumination to ultrahigh-speed CCDs can thus solve the problem of a
lack of incident light. An n- epitaxial layer/p- epitaxial layer/p+ substrate structure was created to collect electrons
generated at the back side traveling to the collection gate. When a photon reaches the deep position near the CCD
memory in the p-well, an electron generated by photoelectric conversion directly mixes into the CCD memory. This
mixing creates noise, making it necessary to reduce the reach of the incident light. Setting the thickness of a double
epitaxial layer to 30 μm, however, will inhibit the generation of this noise. A potential profile for the n-/p-/p+ structure
was calculated using a three-dimensional semiconductor device simulator. The transit time from electron generation to
arrival at the collection gate was also calculated. The concentrations of the n- and p- epitaxial layers were optimized to
minimize transit time, which was ultimately 1.5 ns. This value is adaptive to a frame rate of 100 million frames/second.
Charge transfer simulation of a part of the pixel was conducted to confirm the smooth transfer of electrons without their
staying too long in one place.
We developed an ultrahigh-speed color video camera that operates at 1,000,000 fps (frames per second) and had capacity to store 288 frame memories.
In 2005, we developed an ultrahigh-speed, high-sensitivity portable color camera with a 300,000-pixel single CCD (ISIS-V4: In-situ Storage Image Sensor, Version 4). Its ultrahigh-speed shooting capability of 1,000,000 fps was made possible by directly connecting CCD storages, which record video images, to the photodiodes of individual pixels. The number of consecutive frames was 144. However, longer capture times were demanded when the camera was used during imaging experiments and for some television programs.
To increase ultrahigh-speed capture times, we used a beam splitter and two ultrahigh-speed 300,000-pixel CCDs. The beam splitter was placed behind the pick up lens. One CCD was located at each of the two outputs of the beam splitter. The CCD driving unit was developed to separately drive two CCDs, and the recording period of the two CCDs was sequentially switched. This increased the recording capacity to 288 images, an increase of a factor of two over that of conventional ultrahigh-speed camera.
A problem with the camera was that the incident light on each CCD was reduced by a factor of two by using the beam splitter. To improve the light sensitivity, we developed a microlens array for use with the ultrahigh-speed CCDs. We simulated the operation of the microlens array in order to optimize its shape and then fabricated it using stamping technology. Using this microlens increased the light sensitivity of the CCDs by an approximate factor of two.
By using a beam splitter in conjunction with the microlens array, it was possible to make an ultrahigh-speed color video camera that has 288 frame memories but without decreasing the camera's light sensitivity.
In order to investigate the unsteady flow field around a spiked body in supersonic flow, time-resolved color schlieren
visualization was applied using a high-speed video camera which could take up to 1 000 000 frames per second at full
frame resolution. Conically and spherically tipped spikes of six different lengths could be attached at the center of the
model and their effect on the flow unsteadiness was visually observed. The obtained images revealed in great detail the
interaction between the incoming free stream flow and the high-pressure region near the model base, which could make
its presence known upstream at the tip of the spike by means of displacing the boundary layer on the spike and
subsequently inducing a large-scale instability of the flow.
We are advancing the development of ultrahigh-speed, high-sensitivity CCDs for broadcast use that are capable of capturing smooth slow-motion videos in vivid colors even where lighting is limited, such as at professional baseball games played at night. We have already developed a 300,000 pixel, ultrahigh-speed CCD, and a single CCD color camera that has been used for sports broadcasts and science programs using this CCD. However, there are cases where even higher sensitivity is required, such as when using a telephoto lens during a baseball broadcast or a high-magnification microscope during science programs. This paper provides a summary of our experimental development aimed at further increasing the sensitivity of CCDs using the light-collecting effects of a microlens array.
This paper presents further results of an ongoing experimental and numerical investigation into the unsteady process of
blast wave reflection from straight smooth surfaces. It is shown that basic blast wave phenomena such as the transition
from regular to irregular wave reflection can be adequately and conveniently studied in a laboratory environment by
using small charges with masses in the milligram range. While the laboratory scale generally provides greater
accessibility, it also imposes more stringent conditions on the diagnostics than the large-scale environment. The paper
reviews the previously found considerable discrepancies between numerical and experimental results for the location x<sub><i>tr</i></sub>
of the transition from regular to irregular wave reflection. These are caused by the initially minuscule size and gradual
growth of the Mach stem and the limited resolution of the recording material. Different techniques are used to improve
the accuracy of the experimental determination of the transition point, and a new combination of modern high-speed
photography with the traditional soot technique is shown to be the most promising tool for this purpose.
We have developed an ultrahigh-speed, high-sensitivity portable color camera with a new 300,000-pixel single CCD.
The 300,000-pixel CCD, which has four times the number of pixels of our initial model, was developed by seamlessly
joining two 150,000-pixel CCDs. A green-red-green-blue (GRGB) Bayer filter is used to realize a color camera with the
single-chip CCD. The camera is capable of ultrahigh-speed video recording at up to 1,000,000 frames/sec, and small
enough to be handheld. We also developed a technology for dividing the CCD output signal to enable parallel, highspeed
readout and recording in external memory; this makes possible long, continuous shots up to 1,000 frames/second.
As a result of an experiment, video footage was imaged at an athletics meet. Because of high-speed shooting, even
detailed movements of athletes' muscles were captured. This camera can capture clear slow-motion videos, so it enables
previously impossible live footage to be imaged for various TV broadcasting programs.
We are developing an ultrahigh-speed, high-sensitivity broadcast camera that is capable of capturing clear, smooth slow-motion videos even where lighting is limited, such as at professional baseball games played at night. In earlier work, we developed an ultrahigh-speed broadcast color camera<sup>1</sup>) using three 80,000-pixel ultrahigh-speed, highsensitivity CCDs<sup>2</sup>). This camera had about ten times the sensitivity of standard high-speed cameras, and enabled an entirely new style of presentation for sports broadcasts and science programs. Most notably, increasing the pixel count is crucially important for applying ultrahigh-speed, high-sensitivity CCDs to HDTV broadcasting. This paper provides a summary of our experimental development aimed at improving the resolution of CCD even further: a new ultrahigh-speed high-sensitivity CCD that increases the pixel count four-fold to 300,000 pixels.
An image sensor for an ultra-high-speed video camera was developed. The maximum frame rate, the pixel count and the number of consecutive frames are 1,000,000 fps, 720 x 410 (= 295,200) pixels, and 144 frames. A micro lens array will be attached on the chip, which increases the fill factor to about 50%. In addition to the ultra-high-speed image capturing operation to store image signals in the in-situ storage area adjacent to each pixel, standard parallel readout operation at 1,000 fps for full frame readout is also introduced with sixteen readout taps, for which the image signals are transferred to and stored in a storage device with a large capacity equipped outside the sensor. The aspect ratio of the frame is about 16 : 9, which is equal to that of the HDTV format. Therefore, a video camera with four sensors of the ISIS-V4, which are arranged to form the Bayer’s color filter array, realizes an ultra-high-speed video camera of a semi-HDTV format.
We developed an ultrahigh-speed, high-sensitivity, color camera that captures moving images of phenomena too fast to be perceived by the human eye. The camera operates well even under restricted lighting conditions. It incorporates a special CCD device that is capable of ultrahigh-speed shots while retaining its high sensitivity. Its ultrahigh-speed shooting capability is made possible by directly connecting CCD storages, which record video images, to photodiodes of individual pixels. Its large photodiode area together with the low-noise characteristic of the CCD contributes to its high sensitivity. The camera can clearly capture events even under poor light conditions, such as during a baseball game at night. Our camera can record the very moment the bat hits the ball.