• 한국컴퓨터정보학회논문지
• /
• 제26권6호
• /
• pp.1-8
• /
• 2021

본 논문은 저해상도 적외선영상을 사양이 낮은 임베디드 시스템에서 추론 가능하도록 강화된 VGG 스타일과 Global Average Pooling 조합으로 정확도를 증가시키면서 연산량을 최소화하는 딥러닝 컨볼루션 신경망을 이용한 저해상도 적외선 표적 분류 방법을 제안한다. 제안한 알고리즘은 OKTAL-SE로 생성한 합성영상 클래스 9개 3,723,328개를 분류하였다. 최초 임베디드 추론 가능하도록 파라메터 수가 최소화된 최대풀링 레이어 기준 입력단 8개와 출력단 8개 조합에 비해 강화된 VGG 스타일을 적용한 입력단 4개와 출력단 16개 필터수 조합을 이용하여 연산량은 약 34% 감소시켰으며, 정확도는 약 2.4% 증가시켜 최종 정확도 96.1%을 획득하였다. 추가로 C 코드로 포팅하여 수행시간을 확인하였으며, 줄어든 연산량 만큼 수행 시간이 약 32% 줄어든 것을 확인할 수 있었다.

• 한국컴퓨터정보학회논문지
• /
• 제14권1호
• /
• pp.135-143
• /
• 2009

최근 모바일 장치에서 고해상도 영상의 인코딩 및 디코딩에 대한 요구가 늘어남에 따라 효율적인 영상 코덱 개발의 필요성이 증대되고 있다. 본 논문은 JPEG 디코딩 과정에서 IDCT 변환과 컬러변환 배열간의 선형성을 바탕으로 이들 연산순서를 재배열함으로써 컬러변환 과정에서 요구되는 계산 횟수를 줄이고 재배열된 부동소수점 연산에 정수 맵핑을 적용하여 시간 복잡도를 줄임으로써 실행시간을 크게 단축하는 컬러변환 기법을 제안한다. 또한, 제안된 기법은 연산 재배열 및 정수 맵핑의 양자화오류로 인한 화질 저하를 다중 모드 채도 재구성 기법을 적용하여 보상하도록 한다. 임베디드 시스템 개발 플랫폼에서의 성능평가를 통해 제안 된 기법이 기존의 컬러변환 기법들과 비교하여 복원 영상의 화질 저하를 최소화하면서 실행시간을 크게 단축함을 알 수 있었다.

• 대한토목학회논문집
• /
• 제43권4호
• /
• pp.511-523
• /
• 2023

철근콘크리트 손상 감지를 위한 무인항공기와 딥러닝 연계에 대한 연구가 활발히 진행 중이다. 컨볼루션 신경망은 객체 분류, 검출, 분할 모델의 백본으로 모델 성능에 높은 영향을 준다. 사전학습 컨볼루션 신경망인 모바일넷은 적은 연산량으로 충분한 정확도가 확보 될 수 있어 무인항공기 기반 실시간 손상 감지 백본으로 효율적이다. 바닐라 컨볼루션 신경망과 모바일넷을 분석 한 결과 모바일넷이 바닐라 컨볼루션 신경망의 15.9~22.9% 수준의 낮은 연산량으로도 6.0~9.0% 높은 검증 정확도를 가지는 것으로 평가되었다. 모바일넷V2, 모바일넷V3Large, 모바일넷 V3Small은 거의 동일한 최대 검증 정확도를 가지는 것으로 나타났으며 모바일넷의 철근콘트리트 손상 이미지 특성 추출 최적 조건은 옵티마이저 RMSprop, 드롭아웃 미적용, 평균풀링인 것으로 분석되었다. 본 연구에서 도출된 모바일넷V2 기반 7가지 손상 감지 최대 검증 정확도 75.49%는 이미지 축적과 지속적 학습으로 향상 될 수 있다.

• Journal of Electrical Engineering and Technology
• /
• 제10권4호
• /
• pp.1886-1898
• /
• 2015

In today’s era of advanced technological developments, the threats to the authenticity and integrity of digital images, in a nutshell, the threats to the Image Forensics Research communities have also increased proportionately. This happened as even for the ‘non-expert’ forgers, the availability of image processing tools has become a cakewalk. This image forgery poses a great problem for judicial authorities in any context of trade and commerce. Block matching based image cloning detection system is widely researched over the last 2-3 decades but this was discouraged by higher computational complexity and more time requirement at the algorithm level. Thus, for reducing time need, various dimension reduction techniques have been employed. Since a single technique cannot cope up with all the transformations like addition of noise, blurring, intensity variation, etc. we employ multiple techniques to a single image. In this paper, we have used Fuzzy logic approach for decision making and getting a global response of all the techniques, since their individual outputs depend on various parameters. Experimental results have given enthusiastic elicitations as regards various transformations to the digital image. Hence this paper proposes Fuzzy based cloning detection and classification system. Experimental results have shown that our detection system achieves classification accuracy of 94.12%. Detection accuracy (DAR) while in case of 81×81 sized copied portion the maximum accuracy achieved is 99.17% as regards subjection to transformations like Blurring, Intensity Variation and Gaussian Noise Addition.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권2호
• /
• pp.471-485
• /
• 2023

As data sharing increases explosively, such information encoded in QR code is completely public as private messages are not securely protected. This paper proposes a new 'PROMISE' framework for hiding information based on the QR code projection matrix by using image segmentation without modifying the essential QR code characteristics. Projection matrix mapping, matrix scrambling, fusion image segmentation and steganography with SEL(secret embedding logic) are part of the PROMISE framework. The QR code could be mapped to determine the segmentation site of the fusion image as a binary information matrix. To further protect the site information, matrix scrambling could be adopted after the mapping phase. Image segmentation is then performed on the fusion image and the SEL module is applied to embed the secret message into the fusion image. Matrix transformation and SEL parameters should be uploaded to the server as the secret key for authorized users to decode the private message. And it was possible to further obtain the private message hidden by the framework we proposed. Experimental findings show that when compared to some traditional information hiding methods, better anti-detection performance, greater secret key space and lower complexity could be obtained in our work.

• Journal of Multimedia Information System
• /
• 제4권1호
• /
• pp.43-50
• /
• 2017

In this paper, we propose a method which detects the nose and face of certain human by using the depth image. The proposed method has advantages of the low computational complexity and the high accuracy even in dark environment. Also, the detection accuracy of nose and face does not change in various postures. The proposed method first locates the locally protruding part from the depth image of the human body captured through the depth camera, and then confirms the nose through the depth characteristic of the nose and surrounding pixels. After finding the correct pixel of the nose, we determine the region of interest centered on the nose. In this case, the size of the region of interest is variable depending on the depth value of the nose. Then, face region can be found by performing binarization using the depth histogram in the region of interest. The proposed method can detect the nose and the face accurately regardless of the pose or the illumination of the captured area.

• 대한전자공학회논문지
• /
• 제27권5호
• /
• pp.775-783
• /
• 1990

In this paper we implemented the fast thinning unit by modifying the pipeline architecture which was proposed by Stanley R. Sternberg. The unit is useful in preprocessing such as image representation and pattern recognition etc. This unit is composed of interface part, local memory part, address generation part, thinning processing part and control part. In thinning processing part, we shortened the thinning part which performed by means of look up table using window mapping table. Thus we improved the weakness of SAP, in which the number of delay pipeline and window pipeline are equal to image column size. Two independent memorys using tri-state buffer enable the two direction flow of address generated by address generation part. This unit avoids the complexity of architecture and has flexibility of image size by means of simple modification of logic bits.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1997년도 추계학술대회 학술발표 논문집
• /
• pp.271-274
• /
• 1997

This paper presents an adaptive quantization of image sequences using the Radial Basis Function Network(RBFN) which classifies interframe image blocks. The clssification algorithm consists of two steps. Blocks are classified into NA(No Activity), SA(Small Activity), VA(Verical Activity), and HA(Horizontal Activity) classes according to edges, image activity and AC anergy distribution. RBFN is trained using the classification results of the above algorithm, which are nonlinear classification features are acquired from the complexity and variability of difference blocks. Simulation result shows that the the adaptive quantization using the RBFN method produced better results better results than that of the sorting and MLP methods.

• Journal of information and communication convergence engineering
• /
• 제10권2호
• /
• pp.194-199
• /
• 2012

This paper proposes a novel scheme of a gray scale fingerprint image for a high-accuracy capacitive sensor chip. The conventional grayscale image scheme uses a digital-to-analog converter (DAC) of a large-scale layout or charge-pump circuit with high power consumption and complexity by a global clock signal. A modified capacitive detection circuit for the charge sharing scheme is proposed, which uses a down literal circuit (DLC) with a floating-gate metal-oxide semiconductor transistor (FGMOS) based on a neuron model. The detection circuit is designed and simulated in a 3.3 V, 0.35 ${\mu}m$ standard CMOS process. Because the proposed circuit does not need a comparator and peripheral circuits, the pixel layout size can be reduced and the image resolution can be improved.

• 한국통신학회논문지
• /
• 제21권12호
• /
• pp.3033-3045
• /
• 1996

In this paper, we propse an object-oriented coding method in low bit-rate channels using pyramid structure and residual image compensation. In the motion estimation step, global motion is estimated using a set of multiresolution images constructed in a pyramid structure. We split an input image into two regions based on the gradient value. Regions with larte motions obtain observation points at low resolution level to guarantee robustness to noise and to satisfy a motion constraint equation whereas regions with local motions such as eye, and lips get observation points at the original resolution level. Local motion variations and intesity variations of an image reconstructed by the golbal motion are compensated additionally by using the previous residual image component. Finally, the model failure (MF) region is compensated by the pyramid mapping of the previous displaced frame difference (DFD). Computer simulation results show that the proposed method gives better performance that the convnetional one in terms of the peak signal to noise ratio (PSNR), compression ratio (CR), and computational complexity.