• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.11
• /
• pp.5555-5567
• /
• 2018

State-of-art instance segmentation networks are successful at generating 2D segmentation mask for region proposals with highest classification score, yet 3D object segmentation task is limited to geocentric embedding or detector of Sliding Shapes. To this end, we propose an amodal 3D instance segmentation network called A3IS-CNN, which extends the detector of Deep Sliding Shapes to amodal 3D instance segmentation by adding a new branch of 3D ConvNet called A3IS-branch. The A3IS-branch which takes 3D amodal ROI as input and 3D semantic instances as output is a fully convolution network(FCN) sharing convolutional layers with existing 3d RPN which takes 3D scene as input and 3D amodal proposals as output. For two branches share computation with each other, our 3D instance segmentation network adds only a small overhead of 0.25 fps to Deep Sliding Shapes, trading off accurate detection and point-to-point segmentation of instances. Experiments show that our 3D instance segmentation network achieves at least 10% to 50% improvement over the state-of-art network in running time, and outperforms the state-of-art 3D detectors by at least 16.1 AP.

• Current Optics and Photonics
• /
• v.6 no.3
• /
• pp.323-331
• /
• 2022

Electro-optic modulator (EOM) takes a vital role in connecting the electric and optical fields. Here, we present a heterogeneously integrated EOM based on the lithium niobate-on-insulator (LNOI) platform. The key modulation waveguide structure is a field-enhanced slot waveguide formed by embedding silicon nanowires in a thin-film lithium niobate (LN), which is different from the previously reported LN ridge or etchless LN waveguides. Based on such slot structure, optical mode field area is reduced and enhanced electric field in the slot region can interact well with LN material with high Electro-optic (EO) coefficient. Therefore, the improvements in both aspects have positive effects on enhancing the modulation performance. From results, the corresponding EOM by adding such modulation waveguide structure achieves better performance, where the key half-wave-voltage-length product (V_𝜋L) and 3 dB EO bandwidth are 1.78 V·cm and 40 GHz under the electrode gap width of only 6 ㎛, respectively. Moreover, Lower V_𝜋L can also be achieved. With these characteristics, such field-enhanced waveguide structure could further promote the development of LNOI-based EOM.

• Journal of Korea Multimedia Society
• /
• v.8 no.11
• /
• pp.1462-1471
• /
• 2005

Recently, the medical image has been digitized by the development of computer science and digitization of the medical devices. There are needs for database service of the medical image and long term storage because of the construction of PACS(picture archiving and communication system) following DICOM(digital imaging communications in medicine) standards, telemedicine, and et al. However, it also caused some kinds of problems, such as illegal reproduction of medical image, proprietary rights and data authentication. In this paper, we propose the new digital watermarking technique for medical image based on public key encryption algorithm for integrity verification. It prevents illegal forgery that can be caused after transmitting medical image data remotely. The watermark is the value of bit-plane in wavelet transform of the original image for certification method of integrity verification. We proposed the embedding regions are randomly chosen considering ROI, and a digital signature is made using hash function of MD5 which input is a secret key. The experimental results show that the watermark embedded by the proposed algorithm can survive successfully in image processing operations and that the watermark's invisibility is good.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.7 no.8
• /
• pp.91-98
• /
• 1999

This paper describes the air flow characteristics inside the throttle valve. Tow-dimensional steady incompressible Navier-Strokes equation are solved numerically with embedding the conceopt of the artificial compressibility and adopting the Baldwin-Lomax turbulence model. With varying the valve opening angles(the Reynolds number )such as 15$^{\circ}$(5000) , 45$^{\circ}$(3000) , 75$^{\circ}$(7000) and 90$^{\circ}$(10000), respectively. tow cases, with a valve shaft and without one, are analysed. The pressure loss between the entrance and exit is severe at 15$^{\circ}$, 100 times as larger as that of 90$^{\circ}$ case, which also depends much on the existece of the valve shaft. The counter rotating vortices are formed over the valve plate with the shaft at only 75$^{\circ}$. They are smally and very large scale in front and back of the valve shaft , respectively. The velocity profiles of 15$^{\circ}$ and 90$^{\circ}$ at the exit are almost symmetric to the horizontal center line, however, the symmetricity is no longer maintained at 45$^{\circ}$ and 75$^{\circ}$ , and in addition, the flow at 75$^{\circ}$ is enforced a lot below center line. The pressure distribution on the walls is largely changed near the valve shaft, and its magnitude becomes great as the valve angle decreases.

• Journal of Korea Multimedia Society
• /
• v.9 no.5
• /
• pp.624-634
• /
• 2006

This paper presents a new watermarking scheme that is robust against geometrical attacks such as translation and rotation. The proposed method is based on the conventional PSADT(Polar Coordinates Shape Adaptive Discrete Transform) method which is an robust watermarking scheme for an arbitrarily-shaped image such as character images. The PSADT method shows perfect robustness against geometrical attack if there is no change in the shape of the image object. However, it cannot be utilized to watermark general rectangular images because of the missing alignment between the watermarked signals in the embedding and extracting side. To overcome this problem we propose a new watermarking scheme that aligns the watermark signal using the image inherent feature, especially corner. Namely the proposed method decides a consistent target region whose shape and position isn't changed by any malicious attack and then embeds the watermark in it using the PSADT method. Experimental results show the robustness of the proposed method against geometrical attacks as well as image compression.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.11
• /
• pp.66-78
• /
• 2015

With the development of bio computing technology, DNA watermarking to do as a medium of DNA information has been researched in the latest time. However, DNA information is very important in biologic function unlikely multimedia data. Therefore, the reversible DNA watermarking is required for the host DNA information to be perfectively recovered. This paper presents a reversible DNA watermarking using least square based prediction error expansion for noncodng DNA sequence. Our method has three features. The first thing is to encode the character string (A,T,C,G) of nucleotide bases in noncoding region to integer code values by grouping n nucleotide bases. The second thing is to expand the prediction error based on least square (LS) as much as the expandable bits. The last thing is to prevent the false start codon using the comparison searching of adjacent watermarked code values. Experimental results verified that our method has more high embedding capacity than conventional methods and mean prediction method and also makes the prevention of false start codon and the preservation of amino acids.

• Proceedings of the Korea Technical Association of the Pulp and Paper Industry Conference
• /
• 2006.06b
• /
• pp.305-309
• /
• 2006

This paper presents and illustrates the application of a growth and yield model that supports both forest and mill side volume and value estimates. Traditional forest stand growth and yield models represent the forest landowner view of yield and economics. Predicted yields are estimates of what one would expect from a procurement cruise, and current stumpage prices are applied to investigate optimum management strategies. Optimum management regimes and rotation ages obtained from the forest side view are unlikely to be economically optimal when viewed from the mill side. The actual distribution of recoverable manufactured product and its value are highly dependent on mill technologies and configurations. Overcoming this limitation of growth and yield computer models necessitates the ability to predict and price the expected manufactured distribution of lumber, lineal meters of veneer, and tonnes of air dried pulp fiber yield. With these embedded models, users of the yield simulator can evaluate the economics of possible/feasible management regimes from both the forest and mill business sides. The simulator is a forest side model that has been modified to produce estimates of manufactured product yields by embedding models for 1) pulpwood chip size class distribution and pulp yield for any kappa number (Schultz and Matney, 2002), 2) a lumber yield and pricing model based on the Best Opening Face model developed by the USDA Forest Service Forest Products Laboratory (Lewis, 1985a and Lewis, 1985b), and 3) a lineal meter veneer model derived from peeler block tests. While the model is strictly applicable to planted loblolly pine (Pinus taeda L.) on cutover site-prepared land in the United States (US) Gulf South, the model and computer program are adaptable to any region and forest type.

• Journal of Korea Multimedia Society
• /
• v.10 no.11
• /
• pp.1427-1438
• /
• 2007

Video watermarking technologies are classified into types of four kinds. The first type is to embed the watermark into a raw video signal and to code the watermarked video signal. Most of video watermarking technologies fall into the category of this type. The second type is to apply watermarking to the coding process, such as block DCT and quantization. The third is to directly embed the watermark into the compressed bitstream itself. Generally, it is referred as labelling rather than watermarking. Finally, the fourth is to embed the water mark into MPEG motion vector. This type has the difficulty in real-time process because of the high complexity and has the blocking effects because of DCT-based on coder. In this paper, we proposed the digital video watermarking that embed the watermark in SPIHT video code for I-frame using motion vector analysis. This method can remove the blocking effect occurred at the DCT-based on coder and obtain video data that has progressive transmission property. The proposed method is to select the region for the watermark embedding in I frame using motion vector estimated from the previous P or B frame. And then, it is to perform DWT and embed the watermark based on HVS into the wavelet coefficients in the same subband of DWT as the motion vector direction. Finally, the watermarked video bitstream is obtained by the SPIHT coder. The experimental results verified that the proposed method has the invisibility from the objective and subjective image quality and the robustness against the various SPIHT compression and MPEG re-code.