We attempt the problem of autonomous surveillance for person re-identification. This is an active research area, where most recent work focuses on the open challenges of re-identification, independently of prerequisites of detection and tracking. In this paper, we are interested in designing a complete surveillance system, joining all the pieces of the puzzle together. We start by collecting our own dataset from multiple cameras. Then, we automate the process of detection and tracking of human subjects in the scenes, followed by performing the re-identification task. We evaluate the recognition performance of our system, report its strengths, discuss open challenges and suggest ways to address them.
The manner of holding a pen is distinctive among people. Therefore, pen holding style is useful for person
authentication. In this paper, we propose a biometric person authentication method using features extracted
from images of pen holding style. Images of the pen holding style are captured by a camera, and several features are extracted from the captured images. These features are compared with a reference dataset to calculate dissimilarity scores, and these scores are combined for verification using a three-layer perceptron. Preliminary experiments were performed by using a private database. The proposed system yielded an equal error rate (EER) of 2.6%.
Cancelable approaches for biometric person authentication have been studied to protect enrolled biometric data, and several algorithms have been proposed. One drawback of cancelable approaches is that the performance is inferior to that of non-cancelable approaches. As one solution, we proposed a scheme to enhance the performance of a cancelable approach for online signature verification by combining scores calculated from two transformed datasets generated using two keys. Generally, the same verification algorithm is used for transformed data as for raw (non-transformed) data in cancelable approaches, and, in our previous work, a verification system developed for a non-transformed dataset was used to calculate the scores from transformed data. In this paper, we modify the verification system by using transformed data for training. Several experiments were performed by using public databases, and the experimental results show that the modification of the verification system improved the performances. Our cancelable system combines two scores to make a decision. Several fusion strategies are also considered, and the experimental results are reported here.
Authentication of individuals is rapidly becoming an important issue. The authors previously proposed a Pen-input online signature verification algorithm. The algorithm considers a writer’s signature as a trajectory of pen position, pen pressure, pen azimuth, and pen altitude that evolve over time, so that it is dynamic and biometric. Many algorithms have been proposed and reported to achieve accuracy for on-line signature verification, but setting the threshold value for these algorithms is a problem. In this paper, we introduce a user-generic model generated by AdaBoost, which resolves this problem. When user- specific models (one model for each user) are used for signature verification problems, we need to generate the models using only genuine signatures. Forged signatures are not available because imposters do not give forged signatures for training in advance. However, we can make use of another's forged signature in addition to the genuine signatures for learning by introducing a user generic model. And Adaboost is a well-known classification algorithm, making final decisions depending on the sign of the output value. Therefore, it is not necessary to set the threshold value.
A preliminary experiment is performed on a database consisting of data from 50 individuals. This set consists of western-alphabet-based signatures provide by a European research group. In this experiment, our algorithm gives an FRR of 1.88% and an FAR of 1.60%. Since no fine-tuning was done, this preliminary result looks very promising.
The operation principle and configuration of the Scanning Near-field Optical/Atomic Force Microscope (SNOAM) is introduced in the paper. DNA molecules were imaged in AFM mode and in SNOM mode. The topography images and the fluorescence images of single DNA molecule were obtained. The topography image in SNOM mode is of high resolution. The near-field fluorescence image shows the fluorescence distribution of DNA molecules. (lambda) DNA Molecules, in which YOYO-1 was intercalated, were imaged and characterized. For (lambda) DNA with 5 (mu) M YOYO-1, there is variation in the fluorescence intensity of the DNA and ratios of the fluorescence intensity showed almost integers in each region. As the fluorescence intensity correlated with the area of cross section in the DNA topography, it was suggested that YOYO-1 intercalated in the DNA homogeneously. Contrary, the fluorescence intensity of (lambda) DNA with 500 (mu) M YOYO-1 was heterogeneous and did not correlate with the area of topographic cross section. This suggested that YOYO-1 was not intercalated to (lambda) DNA uniformly in the concentration and intercalated partially and cooperatively.
Conformation change of DNA under binding a DNA binding protein was studied using an AFM. An originally designed AFM mounted on an inverse microscope was used for imaging DNA. High mobility group (HMG) protein and pUC118 DNA were used as DNA binding protein and DNA for investigation in this study. The pUC118 DNA and its mixture with HMG2BJ were imaged by the AFM. In the AFM image of pUC118, DNA strands showed open structure with a few helix sites. The height of the helix part is 1.5 times higher than the ordinary DNA strand. Centrally AFM images of the mixture showed HMG-like particles on DNA strands. The height of the particle is 2-3 times higher than the DNA strands which showed the particles were not aggregating parts but HMG. The HMG bound on a crossing part of DNA which make a loop.
A novel pipette SNOM/AFM probe has been developed for its simple fabrication and applicability to wide wavelength range. The pipette probe is simply fabricated by a successive process of pulling, bending with a CO2 laser, making a hole and coating with a metal layer. The hole is made on the tube at the back side of the tip for applying a light. The pipette probe is mounted on the SNOM/AFM system which includes a confocal microscope. The light is introduced to the hole directly by focusing from a confocal microscope or through an optical fiber. The probe provided clear topographic and optical images for a sample of a patterned chromium layer on a glass substrate and fluorescence beads. A confocal images was also obtained in a wide area of the same samples.
Proc. SPIE. 2836, Chemical, Biochemical, and Environmental Fiber Sensors VIII
KEYWORDS: Near field scanning optical microscopy, Luminescence, Green fluorescent protein, Near field optics, Optical fibers, Biomedical optics, Atomic force microscopy, Microscopy, Spectrographs, Chromium
We have developed scanning near-field optical/atomic force microscopy (SNOM/AFM). The SNOM/AFM uses a bent optical fiber simultaneously as a dynamic force AFM cantilever and a SNOM probe. Resonant frequency of the optical fiber cantilever is 15 - 40 kHz. Optical resolution of the SNOM/AFM images shows less than 50 nm. The SNOM/AFM system contains photon counting system and polychrometer/ICCD system to observe fluorescence image and spectrograph of micro areas, respectively. A SNOM-AFM system was newly applied to analyses of biological samples. In this system a feedback signal from AFM in the noncontact mode was used to scan the probe tip along the surface contour of the sample. An optical fiber with a sharp tip on one end was bent for use as cantilever, and ac amplitude of the cantilever deflection was held constant during scanning by moving the stage. Green fluorescent protein (GFP) absorbs blue light and emits green light. GFP should be a convenient indicator of transformation and one that could allow cells to be separated with fluorescence-activated cell sorting. The gene coding to GFP was cloned in recombinant E.coli and plant cells. Spatial distribution of GFP gene expression was clarified using a SNOM-AFM system. Fluorescent spectroscopic analyses supported GFP was surely produced in E.coli and plant cells. Applications to gene identification in human genomes were also discussed.
KEYWORDS: Near field optics, Near field scanning optical microscopy, Optical fibers, Photoresist materials, Optical signal processing, Image resolution, Near field, Silicon, Semiconducting wafers, Atomic force microscopy
This paper describes the design and applications to optical processing and recording of a Scanning Near-field Optic/Atomic-force Microscope (SNOAM). A sharpened and bent optical fiber was used as a near-field optical probe as well as an atomic force microscope probe in a vertical vibrating mode. SNOAM provides simultaneous topographical and optical images with high resolution beyond the diffraction limit. As an example of an application to optical processing, near-field exposures have been demonstrated by a SNOAM. We produced pit and line patterns exposed and developed in commercial photoresist film. In the processing mode, the pit and line patterns down to a width of 100 nm have been fabricated on a Si wafer through the Integrated Circuits process.
The second-order nonlinear optical coefficients d33 (-2(omega) :(omega) ,(omega) ) of polymer films doped with novel disazo dyes, that have a perfluoroalkylsulfonyl group as an electron acceptor and an extended (pi) - electron system, are determined by second-harmonic generation (SHG) measurements at wavelengths of 900 - 1550 nm. Thin polycarbonate (PC) film doped with 4-(4-(4- perfluorobutylsulfonylphenylazo) naphthylazo)-N-ethyl-N-hydroxyethylaniline showed a large d33 value, 67 pm/V, at a wavelength of 1064 nm. The decay of the d33 value was slow due to the stability of the molecular arrangement in a polymer matrix at high temperature. The large d33 values and the stability of polymer films doped with novel disazo dyes in this study are discussed in terms of electron acceptor substitution and an extended (pi) -electron system.