Frame Memory Research Articles

An efficient frame memory interface of MPEG-2 video encoder ASIC chip

This paper presents an efficient frame memory interface of MPEG-2 video encoder which is accomplished in not only reducing interface buffer size through efficient memory map organization and access timing schedules but also avoiding unnecessary small size buffers and simplifying their control circuits. In this design, 0.5 /spl mu/m CMOS TLM (triple layer metal) standard cells are used as design libraries, and VHDL simulator and logic synthesis tools are used for hardware design and verification, and the hardware emulator that is a C-language model of the proposed architecture is exploited for various test vector generation and functional verification. The improved frame memory interface module takes about 58% less hardware area than the previous design (Kim et al. 1997), and results in reducing the total hardware area of the video encoder ASIC chip up to 24.3%. We also reduced the random memory accesses to save the power consumption caused by the transition of the system-level I/O buses.

A cost effective D-TV system chip set

This paper proposes a cost-effective D-TV system chip set. The video decoder is applicable up to MP@HL, and a new down-decode technique is adopted to reduce the amount of required video frame memory down to a quarter.

並列計算モデルＢｅＮｅｔに基づくプログラミング環境を利用した実時間ロボット視覚の実現

In this paper, we propose to develop a modular real-time robot vision system in a programming environment based on BeNet. BeNet is a parallel computational model for describing real-time systems completely and briefly by software as a set of computational modules of which roles are distinct. The real-time behavior of BeNet does not depends on the hardware and is easy to predict because each module changes the state and the output at regular intervals. An environment in which one can describe a BeNet and implement it immediately on a multi-processor system with image frame memories, makes it possible to develop a large-scale modular system by integrating diverse methods for computer vision in a short term. We present how to design and develop a system that can find an object and track it in real-time based on the top down input of the attention in our environment. We realized a network of modules for feature extraction, object extraction and attention control, and confirmed that the BeNet shows a desirable behavior in the real world and that it is applicable to an autonomous robot.

Time-Resolved High-Resolution Electron Microscopy of Clusters, Surfaces, and Interfaces

Nanoscale materials show a variety of interesting physical and chemical properties. Since the properties are mostly determined by mesoscopic- and atomicscale areas of the materials, characterization of nanoscale materials at the atomic level is crucial for understanding the origin of their properties as well as for tailoring them to industrial needs. High-resolution electron microscopy (HREM) is an effective method to observe atomic structures in nanophase materials. The “structure-imaging technique,” originated by Ueda et al., and Cowley and Iijima, contributed much to the development of the method. Innovations are anticipated in (1) three-dimensional analysis of nanophase materials in real space, (2) observation of dynamic phenomena, and (3) extraction of compositional and physical information from nanometer-sized areas. Using the structure-imaging technique, one can observe directly nanometer-scale particles and defect structures such as voids and line and plane defects in crystals, including surfaces and interfaces. There were previously some limitations on the study of the dynamic process on the atomic level since images were captured on films. The recent development of high-sensitivity silicon-integrated-target-television (SIT-TV) cameras has initiated a new stage of HREM.Our time-resolved-high-resolution-electron-microscopy (TRHREM) system is composed of a 200-kV HREM (JEOL: JEM-2010) equipped with a high-sensitivity SIT-TV camera, various types of dedicated sample holders, an ultrahigh-vacuum (UHV) sample-preparation chamber, and a digital videotape recorder, as partly illustrated in Figure 5. The HREM images at 0.8–1.0-million magnification are recorded on the videotape recorder. In the U.S.-Japan TV system, one “frame image” of 1/30-s time resolution is composed of two interlaced “field images” of 1/60-s time resolution. The digital videotape recorder enables us to output the two “field images” separately from two frame memories.

Diagnostic ultrasound apparatus

A diagnostic ultrasound apparatus includes an ultrasound probe for transmitting and receiving ultrasound beams to and from an object to be observed; a frame memory for storing B-mode ultrasound image data of the object obtained from the ultrasound beams received by the ultrasound probe; a monitor for displaying at least a B-mode ultrasound image of the object formed on the basis of the B-mode ultrasound image data stored in the frame memory; a sampling line setting circuit for freely setting a desired course and range of at least one data sampling line for the B-mode image displayed on the monitor; and a read-out circuit for selectively reading out, from the ultrasound image data stored in the frame memory, ultrasound image data corresponding to the data sampling line. The ultrasound image data which has been read out by the read-out circuit is displayed on the monitor as an M-mode image. According to the diagnostic ultrasound apparatus, the data sampling line is established for any desired course and range unrelated to the forward direction of the ultrasound beams which form the B-mode image displayed on the monitor, and the ultrasound image data which corresponds to the data sampling line is selectively read out from the ultrasound image data of the B-mode image stored in the frame memory. The thus-obtained ultrasound image data of the data sampling line of the desired course, that is the M-mode image is displayed on the monitor. Therefore, observation and diagnosis of the state of motion of any specific part of the patient can be carried out freely in any direction irrespective of the forward direction of the ultrasound beams from the probe.

Optimization Method of Memory Characteristics in Antiferroelectric Liquid Crystals

The memory characteristics of antiferroelectric liquid crystals were studied. The investigation of holding characteristics upon application of a rectangular waveform voltage revealed that hysteresis curve of transmittance versus holding voltage characteristics could be obtained. In the hysteresis range, both the ferroelectric and antiferroelectric states exist stably during a multi-frame periods. The optimization method proposed herein is expected to lead to the development of a display without frame memory.

MPEG2 video decoder and AC-3 audio decoder LSIs for DVD player

We have developed the MPEG2 video decoder and the AC-3 audio decoder LSIs for the DVD player. These decoder LSIs together with single 16 Mbit SDRAM and D/A converters realize the audio/video processing unit with a small number of components for the DVD player. The video decoder LSI adopts the RISC controller to control the dedicated processing hardware cores and a design to integrate the frame memory and bitstream buffers into a single external memory. The audio decoder LSI adopts a 24-bit fixed point DSP core and the dedicated AC-3 accelerator circuit to reduce the amount of software processing in the AC-3 audio decoding.

Real-Time NMR Imaging Systems Using Personal Computers

Two real-time NMR image-processing systems using high-speed personal computers have been developed. The first was made with a MS-DOS (Microsoft Disk Operating System) PC system (CPU, Pentium; clock frequency, 100 MHz) and a homebuilt frame memory board. The second was made with a MS-Windows (Microsoft Windows 95) PC system (CPU, Pentium; clock frequency, 133 MHz). The reconstruction time for one 128 × 128 image was 280 ms for the DOS system and 120 ms for the Windows system, while the image display time was 30 ms for the DOS system and 120 ms for the Windows system. NMR imaging experiments for observing unsteady particle or bubble motion in fluids were performed using these systems. These real-time image-reconstruction systems demonstrate great promise as add-on devices to existing NMR imaging systems.

Multimedia LSI accelerator with embedded DRAM

To succeed in the graphics controllers market, it is important to take advantage of embedded DRAMs, which provide low power consumption, low electromagnetic interference (EMI), smaller board space, and frame memory flexibility (capacity, access speed, and bandwidth). These capabilities benefit portable PC applications in which board space and power consumption are serious considerations. The MSM7680 accelerator is ideally suited for a compact multimedia system because of its smaller, embedded DRAM capacity. The MSM7680 provides high performance and/or a one-chip solution for many graphics display systems. It also integrates the frame buffer with graphics controller functions such as a 2D drawing engine, MPEG-1 decoder, digital/analog converter for RGB analog output, and a clock generator phase-locked loop. The MSM7680 multimedia accelerator uses a 256-bit data bus with embedded 1.25-Mbyte DRAM to increase the rate of data transfers and decrease power consumption.

Very low bit-rate video coding with block partitioning and adaptive selection of two time-differential frame memories

Several studies on very low bit rate video coding have been reported. One of the major goals of the studies is to improve the coding performance, which gives better subjective and objective quality than conventional coding methods at the same bit rate. As the shape and structure of an object in a picture are arbitrary, the performance of traditional coding with block-based motion compensation (MC) is not satisfactory. We present advanced MC schemes for very low bit-rate video coding. The major features of the proposed MC are block-partitioning prediction and the utilization of two time-differential reference frames. This coding scheme improves the image quality around the objects' boundaries and consequently reduces prediction errors. It also works well in the case of object occlusions. The combination of the proposed MC and discrete cosine transformation (DCT) shows a better performance in several test sequences than full-spec H.263.

A bipolar image detector with smart functions

A new type of bipolar image detector has been developed which consists of bipolar sensor cells, bipolar frame memory cells, and signal transfer circuits. The imager with 100 H/spl times/70 V pixels was designed to detect near-infrared LED rays exactly and rapidly, and was fabricated using 2-/spl mu/m Bi-CMOS process. The device, utilizing operations specific to bipolar cells, amplifies signal charges and cancels noises to have a high signal-to-noise ratio (S/N) immune to a long time hold. It also offers several novel functions such as cancellation of external light signal, preliminary fast readout, and main readout of a selected part. Equipped with peripheral circuits, the chip is 3.66/spl times/3.7 mm in size and is very suitable for image processing use.

A 1.6-GB/s data-rate 1-Gb synchronous DRAM with hierarchical square-shaped memory block and distributed bank architecture

This paper describes key techniques for a 1.6-GB/s high data-rate 1-Gb synchronous DRAM (SDRAM). Its high data transfer rate and large memory capacity are targeted to the advanced unified memory system in which a single DRAM (array) is used as both the main memory and the three-dimensional (3-D) graphics frame memory in a time sharing fashion. The 200-MHz high-speed operation is achieved by the unique hierarchical square-shaped memory block (SSMB) layout and the novel distributed bank (D-BANK) architecture. A 0.29 /spl mu/m/sup 2/ cell and 581.8 mm/sup 2/ small die area are achieved using 0.15-/spl mu/m CMOS technology. The /spl times/61 chip uses 196-pin BGA type chip-scale-package (CSP). Implementation of a built-in self-test (BIST) circuit with a margin test capability is also described.

Shack Hartmann wave-front measurement with a large F-number plastic microlens array

A new plastic microlens array, consisting of 900 lenslets, has been developed for the Shack Hartmann wave-front sensor.The individual lens is 300 µm × 300µm and has a focal length of 10 mm, which provides the same focal size, 60 µm in diameter, with a constant peak intensity. One can improve thewave-front measurement accuracy by reducing the spot centroiding error by averaging a few frame memories of an image processor. A deformable mirror for testing the wave-front sensor gives anappropriate defocus and astigmatism, and the laser wave front is measured with a Shack Hartmann wave-front sensor. The measurement accuracy and reproducibility of our wave-front sensor are better than λ/20 and λ/50 (λ = 632.8 nm),respectively, in rms.

Implementation of Experimental Three-Axis Tactile Sensor System Applying Optical Waveguide Plate.

The present paper describes the implementation of and experiments on a three-axis tactile sensor applying an optical wave guide plate. The sensor is comprised of a sparse array of columnar feelers and a dense array of conical feelers, which are formed on opposite sides of a silicon rubber sheet. The columnar feelers contact objects while the conical feelers contact the acrylic optical waveguide plate. Three components of a force vector applied to the columnar feeler are identified from contact areas of four conical feelers which are covered by one columnar feeler. Contact points of conical feelers photographed using a CCD camera are transferred into a frame memory board installed in a personal computer and they are memorized as 8-bit patterns. A one-bit pattern is quantized from the 8-bit pattern by utilizing a suitable threshold value; the contact areas of conical feelers are obtained from the total number of picture cells of the one-bit pattern. In evaluation experiments, several combinations of vertical and horizontal forces are applied to the columnar feeler through a probe attached to a three-axis force sensor. The experimental results show that the three components of the force vector can be calculated from the four contact areas, and that hard material is not favorable for the conical feelers but is suitable for columnar feelers.

A method for hierarchical subband HDTV splitting

An investigation is made on the compatibility of the video formats, standard TV (625/50/2:1/4:3), EDTV (625/50/1:1/16:9), HDTVint (1250/50/2:1/16:9) and HDTVpro (1250/50/1:1/16:9) in three-dimensional subband coding. A hierarchical method is proposed leading to a more reasonable joint solution for the issues of digital transmission evolution and interlace-to-progressive scanning conversion in the frame work of bit rate compression. Conventional HDTVint with the diamond shaped spectrum is proposed to be replaced by a kind of "switching" HDTVint signal which consists of three components and motion detection bits (MDB). The first component and the MDB represent TV. The first plus the second component and the MDB represent EDTV. HDTVpro is split into switching HDTVint and an extra component. The switching threshold and the length of temporal filter banks are discussed. Rearranging the pixels in subbands is recommended. Based on the proposed method, the transmission system can evolve through four steps: TV, EDTV, HDTVint, and HDTVpro. Results of bit rate compression are also presented. TV, EDTV, HDTVint, and HDTVpro are compressed to approximately 40, 80, 128, and 256 Mb/s, respectively. Interlace scanning has drawbacks such as line flicker and line crawling as the eye follows the line structure in the picture. Progressive version switching TV and HDTVint could be reconstructed using one frame memory without the problems of handling temporal-vertical diamond shaped spectra of conventional TV and HDTVint based on one-frame memory algorithm in the receiver.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

画素間引きを用いた超高精細映像ハンドリングシステムの構成と動画生成への応用

We developed an image-handling system for super-high-resolution images of 3840 x 2070 pixels. In this system, each image is divided into four images using a selection circuit that subsamples in the horizontal and vertical directions. The circuit is placed between the CPU data bus and the frame memory, and the images are distributed to four HDTV frame memories. This allows the CPU to handle one continuous image space, and also permits the use of HDTV techniques and equipment in system construction. We also developed a motion-picture production system using this system, and evaluated the overall spatial frequency characteristics of a 35mm motion picture. Resolution of more than 1600 TV-lines was obtained from an 8-perforation per frame negative film.

特集 民生用カメラ技術 各種機能の現状 画像メモリー応用

特集 民生用カメラ技術 各種機能の現状 画像メモリー応用

Automated Fatigue Crack Growth Tracking System with Image Processor.

An automatic fatigue crack tracking system with an image processor has been newly developed. This system consists of a microscope with a CCD camera, an image processor, a position measuring system, an automatic stage which moves the microscope, a personal computer and a fatigue testing machine. The tracking procedures fall into the following four distinct steps. Step 1 : When a maximum load is applied to the cracked specimen, the system obtains the image around the crack tip and stores the image to the frame memory. Step 2 : After appropriate stress cycles, the system obtains the image at the same position again. Step 3 : The image processor freezes the motion and subtracts the stored image from the image obtained in Step 2. Then the subtracted image is emphasized by adding the offset brightness level. Step 4 : The dark image which is obtained by preceding procedures, is tracked with a template by the position measuring system. The amount of shift of the template represents the length of fatigue crack extension. The system moves the microscope by a distance equal to the length of the crack extension. We applied this system to the fatigue crack propagation tests on steel and Si3N4 ceramics, and confirmed the reliability of our data.

Ultrasonic imaging system capable of displaying 3-dimensional angiogram in real time mode

In an ultrasonic imaging system, an ultrasonic angiographic image of a blood vessel, equivalent to an X-ray angiographic image, is produced in a real time by utilizing an IIR digital filter with a ROM and a frame memory. The ultrasonic imaging system comprises: a scanning unit for scanning a biological body under medical examination within a three-dimensional scanning plane involving the biological body by utilizing ultrasonic pulses to obtain echo signals produced from echo pulses reflected from the biological body; an ultrasonic image data producing unit for producing a plurality of ultrasonic scanned image data about the scanned biological body with regard to the respective scanning planes in response to the echo signals; a bloodflow-distribution image data producing unit for producing a plurality of bloodflow-distribution image data about the scanned biological body with regard to the respective scanning planes in response to the echo signals; a storage unit for temporarily storing the plurality of bloodflow-distribution image data, while sequentially updating and superimposing the plural bloodflow-distribution image data with each other during a predetermined scanning period.

Ultrasonic Doppler blood flow measuring apparatus

An ultrasonic Doppler blood flow measuring apparatus comprises a plurality of frame memories for storing information pieces about a blood flow speed in a unit of frame, circuits for comparing blood flow speed information pieces at corresponding addresses on the plurality of frame, memories and determining a speed difference, a circuit for comparing the speed difference with a predetermined threshold value and determining a blood flow at the corresponding address as an arterial flow when the speed difference exceeds the threshold value but as a venous flow when the speed difference is below the threshold value, and a circuit for displaying the determined arterial and venous flows in different colors.