Traditional fractal pattern design has some disadvantages such as inability to effectively reflect the characteristics of real scenery and texture. We propose a novel pattern design technique combining fractal geometr...Traditional fractal pattern design has some disadvantages such as inability to effectively reflect the characteristics of real scenery and texture. We propose a novel pattern design technique combining fractal geometry and image texture synthesis to solve these problems. We have improved Wei and Levoy (2000)’s texture synthesis algorithm by first using two-dimensional autocorrelation function to analyze the structure and distribution of textures, and then determining the size of L neighborhood. Several special fractal sets were adopted and HSL (Hue, Saturation, and Light) color space was chosen. The fractal structure was used to manipulate the texture synthesis in HSL color space where the pattern’s color can be adjusted conveniently. Experiments showed that patterns with different styles and different color characteristics can be more efficiently generated using the new technique.展开更多
In order to fast transmission and processing of medical images and do not need to install client and plug-ins, the paper designed a kind of medical image reading system based on BS structure. This system improved the ...In order to fast transmission and processing of medical images and do not need to install client and plug-ins, the paper designed a kind of medical image reading system based on BS structure. This system improved the existing IWEB in the framework of PACS client image processing, medical image based on the service WEB completion port model. To realize the fast loading images with high concurrency, compared with the traditional WEB PACS, this system has the advantages of no client without plug-in installation, at the same time in the transmission and processing performance image has been greatly improved.展开更多
ANNs （Artificial neural networks） are used extensively in remote sensing image processing. It has been proven that BPNNs （back-propagation neural networks） have high attainable classification accuracy. However, th...ANNs （Artificial neural networks） are used extensively in remote sensing image processing. It has been proven that BPNNs （back-propagation neural networks） have high attainable classification accuracy. However, there is a noticeable variation in the achieved accuracies due to different network designs and implementations. Hence, researchers usually need to conduct several experimental trials before they can finalize the network design. This is a time consuming process which significantly reduces the effectiveness of using BPNNs and the final design may still not be optimal. Therefore, there is a need to see whether there are some common guidelines for effective design and implementation of BPNNs. With this aim in mind, this paper attempts to find and summarize the common guidelines suggested by different authors through literature review and discussion of the findings. To provide readers with background and contextual information, some ANN fundamentals are also introduced.展开更多
This paper briefly introduces the characteristics and structure of symbol QR two-dimensional code, a detailed analysis of the image processing method to identify QR code of the whole process, and the bilinear mapping ...This paper briefly introduces the characteristics and structure of symbol QR two-dimensional code, a detailed analysis of the image processing method to identify QR code of the whole process, and the bilinear mapping method is applied to image correction, the final steps of decoding are given. The actual test results show that, the design algorithm has theoretical and practical, this recognition system can correctly read QR code, and has high recognition rate and recognition speed, has practical value and application prospect.展开更多
The Image sensor needs various image processing by Image Signal Processor （ISP） to improve image quality. Conventional cameras have their own software ISP functions to perform in PC instead of using commercial ISP c...The Image sensor needs various image processing by Image Signal Processor （ISP） to improve image quality. Conventional cameras have their own software ISP functions to perform in PC instead of using commercial ISP chips. However these methods have problems such as large computation for image processing. In this paper, th authors proposed ISP that significantly reduced chip area by efficiently sharing of hardware and software. Large operation blocks are designed to hardware for high performances, and hardware is imployed simultaneously with software considering the size of the hardware. The implemented ISP can process Video Graphics Array （VGA） （640 ＊ 480） images and has 91 450 gates size in 0. 35 μm process.展开更多
With the help of engineering technologies, there are varieties of medical supports such as surgical simulations, preoperative plans, intraoperative supports, and diagnostic explanations. In order to perform preoperati...With the help of engineering technologies, there are varieties of medical supports such as surgical simulations, preoperative plans, intraoperative supports, and diagnostic explanations. In order to perform preoperative plans, the simulation model of the target organ is necessary, and is usually generated from some fine images obtained by imaging devices such as CT （computed tomography） or MRI （magnetic resonance imaging）. These devices, however, are very large and it is difficult to use them during the surgery although there is a kind of MRI that can be used during operations. On the other hand, there is a device called Ultrasonography （US Echo）. It is so handy and can be easily used during operations without radiation or magnetic effect. US images, however, are not usually used for the construction of 3D model for preoperative plans since there are so many noises in the image, and the quality of the image is not so good compared to the images obtained by CT or MRI. Therefore, the authors have tried to construct 3D volumetric shape model from just two 2D US （ultrasound） images, which are the horizontal and vertical cross-section images of the target organ. Their concrete target organ and surgery are the aorta and AAE （annuloaortic ectasia）, respectively. With some technologies of image processing, the boundary of the aorta is extracted for both cross-section images, and the thickness of the aorta is estimated. These processes are applied to not only the aorta wall but also the aorta valve. Finally, the aorta shape model can be constructed by inserting the estimated horizontal aorta images into the estimated vertical one. The constructed model has three dimensions so that it can be viewed from any point of view.展开更多
基金Project supported by the Natural Science Foundation of Zhejiang Province (No. M603228), Zhejiang Science and Technology Plan Project, and Ningbo Science Foundation for Doctor, China
文摘Traditional fractal pattern design has some disadvantages such as inability to effectively reflect the characteristics of real scenery and texture. We propose a novel pattern design technique combining fractal geometry and image texture synthesis to solve these problems. We have improved Wei and Levoy (2000)’s texture synthesis algorithm by first using two-dimensional autocorrelation function to analyze the structure and distribution of textures, and then determining the size of L neighborhood. Several special fractal sets were adopted and HSL (Hue, Saturation, and Light) color space was chosen. The fractal structure was used to manipulate the texture synthesis in HSL color space where the pattern’s color can be adjusted conveniently. Experiments showed that patterns with different styles and different color characteristics can be more efficiently generated using the new technique.
文摘In order to fast transmission and processing of medical images and do not need to install client and plug-ins, the paper designed a kind of medical image reading system based on BS structure. This system improved the existing IWEB in the framework of PACS client image processing, medical image based on the service WEB completion port model. To realize the fast loading images with high concurrency, compared with the traditional WEB PACS, this system has the advantages of no client without plug-in installation, at the same time in the transmission and processing performance image has been greatly improved.
文摘ANNs （Artificial neural networks） are used extensively in remote sensing image processing. It has been proven that BPNNs （back-propagation neural networks） have high attainable classification accuracy. However, there is a noticeable variation in the achieved accuracies due to different network designs and implementations. Hence, researchers usually need to conduct several experimental trials before they can finalize the network design. This is a time consuming process which significantly reduces the effectiveness of using BPNNs and the final design may still not be optimal. Therefore, there is a need to see whether there are some common guidelines for effective design and implementation of BPNNs. With this aim in mind, this paper attempts to find and summarize the common guidelines suggested by different authors through literature review and discussion of the findings. To provide readers with background and contextual information, some ANN fundamentals are also introduced.
文摘This paper briefly introduces the characteristics and structure of symbol QR two-dimensional code, a detailed analysis of the image processing method to identify QR code of the whole process, and the bilinear mapping method is applied to image correction, the final steps of decoding are given. The actual test results show that, the design algorithm has theoretical and practical, this recognition system can correctly read QR code, and has high recognition rate and recognition speed, has practical value and application prospect.
基金sponsored by ETRI System Semiconductor Industry Promotion Center,Human Resource Development Project for SoC Convergence and“System IC2010”project of Korea Ministry of Knowledge Economy
文摘The Image sensor needs various image processing by Image Signal Processor （ISP） to improve image quality. Conventional cameras have their own software ISP functions to perform in PC instead of using commercial ISP chips. However these methods have problems such as large computation for image processing. In this paper, th authors proposed ISP that significantly reduced chip area by efficiently sharing of hardware and software. Large operation blocks are designed to hardware for high performances, and hardware is imployed simultaneously with software considering the size of the hardware. The implemented ISP can process Video Graphics Array （VGA） （640 ＊ 480） images and has 91 450 gates size in 0. 35 μm process.
文摘With the help of engineering technologies, there are varieties of medical supports such as surgical simulations, preoperative plans, intraoperative supports, and diagnostic explanations. In order to perform preoperative plans, the simulation model of the target organ is necessary, and is usually generated from some fine images obtained by imaging devices such as CT （computed tomography） or MRI （magnetic resonance imaging）. These devices, however, are very large and it is difficult to use them during the surgery although there is a kind of MRI that can be used during operations. On the other hand, there is a device called Ultrasonography （US Echo）. It is so handy and can be easily used during operations without radiation or magnetic effect. US images, however, are not usually used for the construction of 3D model for preoperative plans since there are so many noises in the image, and the quality of the image is not so good compared to the images obtained by CT or MRI. Therefore, the authors have tried to construct 3D volumetric shape model from just two 2D US （ultrasound） images, which are the horizontal and vertical cross-section images of the target organ. Their concrete target organ and surgery are the aorta and AAE （annuloaortic ectasia）, respectively. With some technologies of image processing, the boundary of the aorta is extracted for both cross-section images, and the thickness of the aorta is estimated. These processes are applied to not only the aorta wall but also the aorta valve. Finally, the aorta shape model can be constructed by inserting the estimated horizontal aorta images into the estimated vertical one. The constructed model has three dimensions so that it can be viewed from any point of view.
