2013Research papers

1.Development of Full-Color Laser Head-Up Display

Yasuo Tanahashi, Osamu Kasono, Takuma Yanagisawa, Takayuki Nomoto, Ikuya Kikuchi, Toshiharu Ezuka

Summary

A compact head-up display (HUD) system, which can be attached on a car's ceiling, is developed. Wider field of view (FOV, 17.1° x 5.7°) is achieved in combination with a laser projector, a transparent screen, a field lens and a combiner which is concaved half mirror. Optical see-through AR navigation can be realized with it. This paper is based on a report in IDW/AD'12(1).

Click for Details (PDF 1.7 MB)
in Japanese

2.The practical use and some analysis of car probe data

Kazunori Hashimoto, Takahiro Kamada, Kenji Tsunekawa

Summary

We are developing new service by conducting various analyses from car probe data, what is called "Big Data Processing", and will be expected increasingly from now on. This paper shows the situations of service using car probe data and some examples of the analysis results.

Click for Details (PDF 2.1 MB)
in Japanese

3.Mobile Receiver for Legacy ATSC

Tomoaki Iwai, Yukio Hayashi, Yoshinori Abe

Summary

Despite the common perception that mobile reception of the legacy ATSC signal is nearly impossible, we have succeeded in developing a mobile receiver prototype for the legacy ATSC. Laboratory testing has shown that the prototype works well at speeds over 200 km/h. In this paper, we present the results of a field evaluation conducted in the Los Angeles area. The prototype delivered quasi-error-free video and audio in almost all areas around Los Angeles, and achieved a total reception rate of over 98% to show that the developed technology is immediately applicable to consumer products. After the field evaluation, various analyses were made on the captured data. The results show that the 4-diversity receiver with sufficient estimator tap length can achieve nearly ideal reception where error occurrence is mostly determined by total reception power with little regard to multipath configuration. Delay profile analysis shows that delay spread of the actual field is sometimes wider than 50 microseconds, requiring a channel estimator over 512 taps.

Click for Details (PDF 2.5 MB)
in Japanese

4.Development of Color Tunable and Dimming OLED Lighting

Isamu Ohshita

Summary

We developed the world's first color-tunable and dimming organic light emitting diodes (OLED) having a hole injection layer formed by wet process. Our OLED lighting consists of an OLED panel and a drive/control circuit board. The OLED panel has the light emitting area of 123.1 mm × 123.1 mm, lifetime of around 8,000hours (LT70; time to loss 30% of initial luminance), luminous efficiency of 31 lm/W, and Color Rendering Index Ra of 84 at the white luminance of 1,000 cd/m². In addition, the panel has red, green and blue stripe light emitting layers fabricated by vacuum deposition method, so it can change the emitting color. The drive/control circuit has communication functions compatible with international standard DMX512-A and DALI. Thus, our OLED lighting can provide various lighting effects with an external controller. And also, we developed new models that are an equal-sized model, a small-sized one and a circuit-boardseparated one. We introduced red, yellow and blue light emitting layers into them, so we have realized lifetime of around 8,000hours (LT70) at the white luminance of 2,000 cd/m² and luminous efficiency of 501 m/W at the white luminance of 1,000 cd/m².

Click for Details (PDF 2.8 MB)
in Japanese

5.Solution-Processed Organic Thin-Film Transistor Array for Active-Matrix Organic Light-Emitting Diode

Chihiro Harada, Takuya Hata, Takashi Chuman, Shinichi Ishizuka, Atsushi Yoshizawa

Summary

We developed a 3-inch QQVGA organic thin-film transistor (OTFT) array with an inkjetted organic semiconductor. All layers except electrodes were fabricated by solution processes. The OTFT performed well without hysteresis, and the field-effect mobility in the saturation region was 0.45 cm²/Vs, the threshold voltage was 3.3 V, and the on/off current ratio was more than 106. We demonstrated a 3-inch activematrix organic light-emitting diode (AMOLED) display driven by the OTFT array. The display could provide clear moving image.

Click for Details (PDF 1.7 MB)
in Japanese

6.Traction Control System Based on Adaptive Model Following Control for Electric Vehicles

Masahiro Kato, Dejun Yin (Keio Univ.)

Summary

In order to improve the driving performance of electric vehicles with in-wheel motors, we developed a traction control system that adapts to road conditions. The feedback gain and the time constant of the control system are adaptively generated by the estimated slip ratio and friction coefficient during driving. The simulation and the experimental results show that the control system achieves high acceleration performance in normal road conditions and restrains wheel slip on slippery roads. In addition, we discussed the possibility of applying the traction control method to a probe car system.

Click for Details (PDF 1.9 MB)
in Japanese

7.Mastering of high-density circumferentially aligned patterns and fabrication of nanoimprint molds

Takeru Okada, Jiro Fujimori, Hiroshi Tanaka, Hiroaki Kitahara, Tetsuya Iida

Summary

High-density circumferentially aligned patterns are required for realizing next-generation patterned magnetic media or large-capacity optical disks. In this report, high-density electron-beam lithography process was investigated using an electron-beam recorder with a rotary stage. By changing development condition, substrates, and resists, high-density circumferentially-aligned resist patterns up to 1 Tdots/in.2 were demonstrated. Furthermore, nanoimprint molds including master molds and replica quartz molds were fabricated by applying a liftoff process.

Click for Details (PDF 2.7 MB)
in Japanese

8.A Study on Feature-Extraction Methods for Improvement of Image-Recognition Performance

Toshiaki Inoue

Summary

As rapid spread of various devices with built-in cameras, it is strongly expected to develop advanced image recognition technologies. Robust feature extraction methods such as SIFT and SURF are widely used in many kinds of generic object recognition tasks, and further improvements on these methods are essential to boost recognition performance up effectively. In this paper, we describe an improvement of SIFT on keypoint detection. In original SIFT, DoG filter is used to detect keypoints from input images, whereas in our simple method, fixed window laplacian filter is adopted to improve both detection error and repeatability. Repeatability test results using benchmark datasets show that our proposed method outperforms original SIFT and SURF. Image-verification test results also show that the method is effective to improve discriminative performance.

Click for Details (PDF 1.7 MB)
in Japanese

You will need the latest Adobe Acrobat Reader to view the PDF files on this site.
Click on the icon at left to download your free copy of Adobe Acrobat Reader.