Pub Date : 2007-11-12DOI: 10.1109/ICIP.2007.4379115
Chunhua Chen, Y. Shi, Guorong Xuan
A texture image is of noisy nature in its spatial representation. As a result, the data hidden in texture images, in particular in raw texture images, are hard to detect with current steganalytic methods. We propose an effective universal steganalyzer in this paper, which combines features, i.e., statistical moments of 1-D and 2-D characteristic functions extracted from the spatial representation and the block discrete cosine transform (BDCT) representations (with a set of different block sizes) of a given test image. This novel scheme can greatly improve the capability of attacking steganographic methods applied to texture images. In addition, it is shown that this scheme can be used as an effective universal steganalyzer for both texture and non-texture images.
{"title":"Steganalyzing Texture Images","authors":"Chunhua Chen, Y. Shi, Guorong Xuan","doi":"10.1109/ICIP.2007.4379115","DOIUrl":"https://doi.org/10.1109/ICIP.2007.4379115","url":null,"abstract":"A texture image is of noisy nature in its spatial representation. As a result, the data hidden in texture images, in particular in raw texture images, are hard to detect with current steganalytic methods. We propose an effective universal steganalyzer in this paper, which combines features, i.e., statistical moments of 1-D and 2-D characteristic functions extracted from the spatial representation and the block discrete cosine transform (BDCT) representations (with a set of different block sizes) of a given test image. This novel scheme can greatly improve the capability of attacking steganographic methods applied to texture images. In addition, it is shown that this scheme can be used as an effective universal steganalyzer for both texture and non-texture images.","PeriodicalId":131177,"journal":{"name":"2007 IEEE International Conference on Image Processing","volume":"14 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132057181","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2007-11-12DOI: 10.1109/ICIP.2007.4378905
M. Tico, Markku Vehviläinen
The objective of image stabilization is to prevent or remove the motion blur degradation from images. We introduce a new approach to image stabilization based on combining information available in two differently exposed images of the same scene. In addition to the image normally captured by the system, with an exposure time determined by the illumination conditions, a very shortly exposed image is also acquired. The difference between the exposure times of the two images determines differences in their degradations which are exploited in order to recover the original image of the scene. We formulate the problem as a maximum a posteriori (MAP) estimation based on the degradation models of the two observed images, as well as by imposing an edge-preserving image prior. The proposed method is demonstrated through a series of simulation experiments, and visual examples on natural images.
{"title":"Image Stabilization Based on Fusing the Visual Information in Differently Exposed Images","authors":"M. Tico, Markku Vehviläinen","doi":"10.1109/ICIP.2007.4378905","DOIUrl":"https://doi.org/10.1109/ICIP.2007.4378905","url":null,"abstract":"The objective of image stabilization is to prevent or remove the motion blur degradation from images. We introduce a new approach to image stabilization based on combining information available in two differently exposed images of the same scene. In addition to the image normally captured by the system, with an exposure time determined by the illumination conditions, a very shortly exposed image is also acquired. The difference between the exposure times of the two images determines differences in their degradations which are exploited in order to recover the original image of the scene. We formulate the problem as a maximum a posteriori (MAP) estimation based on the degradation models of the two observed images, as well as by imposing an edge-preserving image prior. The proposed method is demonstrated through a series of simulation experiments, and visual examples on natural images.","PeriodicalId":131177,"journal":{"name":"2007 IEEE International Conference on Image Processing","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132104289","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2007-11-12DOI: 10.1109/ICIP.2007.4378895
E. Akyol, D. Mukherjee, Yuxin Liu
A methodology for complexity scalable video encoding and complexity control within the framework of the H.264/AVC video encoder is presented. To yield good rate-distortion performance under strict complexity/time constraints for instance in real-time communication, a framework for optimal complexity allocation at the macroblock level is necessary. We developed a macroblock level fast motion estimation based complexity scalable motion/mode search algorithm where the complexity is adapted jointly by parameters that determine the aggressiveness of an early stop criteria, the number of ordered modes searched, and the accuracy of motion estimation steps for the INTER modes. Next, these complexity parameters are adapted per macroblock based on a control loop to approximately satisfy an encoding frame rate target. The optimal manner of adapting the parameters is derived from prior training. Results using the developed scalable complexity H.264/AVC encoder demonstrate the benefit of adaptive complexity allocation over uniform complexity scaling.
{"title":"Complexity Control for Real-Time Video Coding","authors":"E. Akyol, D. Mukherjee, Yuxin Liu","doi":"10.1109/ICIP.2007.4378895","DOIUrl":"https://doi.org/10.1109/ICIP.2007.4378895","url":null,"abstract":"A methodology for complexity scalable video encoding and complexity control within the framework of the H.264/AVC video encoder is presented. To yield good rate-distortion performance under strict complexity/time constraints for instance in real-time communication, a framework for optimal complexity allocation at the macroblock level is necessary. We developed a macroblock level fast motion estimation based complexity scalable motion/mode search algorithm where the complexity is adapted jointly by parameters that determine the aggressiveness of an early stop criteria, the number of ordered modes searched, and the accuracy of motion estimation steps for the INTER modes. Next, these complexity parameters are adapted per macroblock based on a control loop to approximately satisfy an encoding frame rate target. The optimal manner of adapting the parameters is derived from prior training. Results using the developed scalable complexity H.264/AVC encoder demonstrate the benefit of adaptive complexity allocation over uniform complexity scaling.","PeriodicalId":131177,"journal":{"name":"2007 IEEE International Conference on Image Processing","volume":"23 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132192617","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2007-11-12DOI: 10.1109/ICIP.2007.4379261
Jyri J. Kivinen, Erik B. Sudderth, Michael I. Jordan
We develop a hierarchical, nonparametric statistical model for wavelet representations of natural images. Extending previous work on Gaussian scale mixtures, wavelet coefficients are marginally distributed according to infinite, Dirichlet process mixtures. A hidden Markov tree is then used to couple the mixture assignments at neighboring nodes. Via a Monte Carlo learning algorithm, the resulting hierarchical Dirichlet process hidden Markov tree (HDP-HMT) model automatically adapts to the complexity of different images and wavelet bases. Image denoising results demonstrate the effectiveness of this learning process.
{"title":"Image Denoising with Nonparametric Hidden Markov Trees","authors":"Jyri J. Kivinen, Erik B. Sudderth, Michael I. Jordan","doi":"10.1109/ICIP.2007.4379261","DOIUrl":"https://doi.org/10.1109/ICIP.2007.4379261","url":null,"abstract":"We develop a hierarchical, nonparametric statistical model for wavelet representations of natural images. Extending previous work on Gaussian scale mixtures, wavelet coefficients are marginally distributed according to infinite, Dirichlet process mixtures. A hidden Markov tree is then used to couple the mixture assignments at neighboring nodes. Via a Monte Carlo learning algorithm, the resulting hierarchical Dirichlet process hidden Markov tree (HDP-HMT) model automatically adapts to the complexity of different images and wavelet bases. Image denoising results demonstrate the effectiveness of this learning process.","PeriodicalId":131177,"journal":{"name":"2007 IEEE International Conference on Image Processing","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130196490","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2007-11-12DOI: 10.1109/ICIP.2007.4379168
Francesc Aulí Llinàs, J. Serra-Sagristà, Joan Bartrina-Rapesta, J. L. Monteagudo-Pereira
Quality scalability is a fundamental feature of JPEG2000, achieved through the use of quality layers. Two points, related with the use of quality layers, may need to be addressed when dealing with JPEG-2000 code-streams: 1) the lack of quality scalability of single quality layer code-streams, and 2) the non rate-distortion optimality of windows of interest transmission. This paper introduces a new rate control method that can be applied to already encoded code-streams, addressing these two points. Its main key-feature is a novel characterization that can fairly estimate the rate-distortion slope of the coding passes of code-blocks without using any measure based on the original image or related with the encoding process. Experimental results suggest that the proposed method is able to supply quality scalability to already encoded code-streams achieving a near-optimal coding performance. The low computational costs of the method makes it suitable for use in interactive transmissions.
{"title":"Enhanced Quality Scalability for JPEG2000 Code-Streams by the Characterization of the Rate-Distortion Slope","authors":"Francesc Aulí Llinàs, J. Serra-Sagristà, Joan Bartrina-Rapesta, J. L. Monteagudo-Pereira","doi":"10.1109/ICIP.2007.4379168","DOIUrl":"https://doi.org/10.1109/ICIP.2007.4379168","url":null,"abstract":"Quality scalability is a fundamental feature of JPEG2000, achieved through the use of quality layers. Two points, related with the use of quality layers, may need to be addressed when dealing with JPEG-2000 code-streams: 1) the lack of quality scalability of single quality layer code-streams, and 2) the non rate-distortion optimality of windows of interest transmission. This paper introduces a new rate control method that can be applied to already encoded code-streams, addressing these two points. Its main key-feature is a novel characterization that can fairly estimate the rate-distortion slope of the coding passes of code-blocks without using any measure based on the original image or related with the encoding process. Experimental results suggest that the proposed method is able to supply quality scalability to already encoded code-streams achieving a near-optimal coding performance. The low computational costs of the method makes it suitable for use in interactive transmissions.","PeriodicalId":131177,"journal":{"name":"2007 IEEE International Conference on Image Processing","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130216360","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2007-11-12DOI: 10.1109/ICIP.2007.4378921
Yang Wu, Nanning Zheng, Qubo You, S. Du
This paper presents a novel, effective way to improve the object recognition performance of a biologically-motivated model by learning informative visual features. The original model has an obvious bottleneck when learning features. Therefore, we propose a circumspect algorithm to solve this problem. First, a novel information factor was designed to find the most informative feature for each image, and then complementary features were selected based on additional information. Finally, an intra-class clustering strategy was used to select the most typical features for each category. By integrating two other improvements, our algorithm performs better than any other system so far based on the same model.
{"title":"Object Recognition by Learning Informative, Biologically Inspired Visual Features","authors":"Yang Wu, Nanning Zheng, Qubo You, S. Du","doi":"10.1109/ICIP.2007.4378921","DOIUrl":"https://doi.org/10.1109/ICIP.2007.4378921","url":null,"abstract":"This paper presents a novel, effective way to improve the object recognition performance of a biologically-motivated model by learning informative visual features. The original model has an obvious bottleneck when learning features. Therefore, we propose a circumspect algorithm to solve this problem. First, a novel information factor was designed to find the most informative feature for each image, and then complementary features were selected based on additional information. Finally, an intra-class clustering strategy was used to select the most typical features for each category. By integrating two other improvements, our algorithm performs better than any other system so far based on the same model.","PeriodicalId":131177,"journal":{"name":"2007 IEEE International Conference on Image Processing","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130293363","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2007-11-12DOI: 10.1109/ICIP.2007.4378878
Rafał K. Mantiuk, Grzegorz Krawczyk, K. Myszkowski, H. Seidel
Vast majority of digital images and video material stored today can capture only a fraction of visual information visible to the human eye and does not offer sufficient quality to fully exploit capabilities of new display devices. High dynamic range (HDR) image and video formats encode the full visible range of luminance and color gamut, thus offering ultimate fidelity, limited only by the capabilities of the human eye and not by any existing technology. In this paper we demonstrate how existing image and video compression standards can be extended to encode HDR content efficiently. This is achieved by a custom color space for encoding HDR pixel values that is derived from the visual performance data. We also demonstrate how HDR image and video compression can be designed so that it is backward compatible with existing formats.
{"title":"High Dynamic Range Image and Video Compression - Fidelity Matching Human Visual Performance","authors":"Rafał K. Mantiuk, Grzegorz Krawczyk, K. Myszkowski, H. Seidel","doi":"10.1109/ICIP.2007.4378878","DOIUrl":"https://doi.org/10.1109/ICIP.2007.4378878","url":null,"abstract":"Vast majority of digital images and video material stored today can capture only a fraction of visual information visible to the human eye and does not offer sufficient quality to fully exploit capabilities of new display devices. High dynamic range (HDR) image and video formats encode the full visible range of luminance and color gamut, thus offering ultimate fidelity, limited only by the capabilities of the human eye and not by any existing technology. In this paper we demonstrate how existing image and video compression standards can be extended to encode HDR content efficiently. This is achieved by a custom color space for encoding HDR pixel values that is derived from the visual performance data. We also demonstrate how HDR image and video compression can be designed so that it is backward compatible with existing formats.","PeriodicalId":131177,"journal":{"name":"2007 IEEE International Conference on Image Processing","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130463663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2007-11-12DOI: 10.1109/ICIP.2007.4379975
B. Davis, T. Ralston, D. Marks, S. Boppart, P. Carney
Optical coherence tomography (OCT) is an optical ranging technique analogous to radar - detection of back-scattered light produces a signal that is temporally localized at times-of-flight corresponding to the location of scatterers in the object. However the interferometric collection technique used in OCT allows, in principle, the coherent collection of data, i.e. amplitude and phase information can be extracted. Interferometric synthetic aperture microscopy (ISAM) adds phase-stable data collection to OCT instrumentation and employs physics-based processing analogous to that used in synthetic aperture radar (SAR). That is, the complex nature of the coherent data is exploited to give gains in image quality. Specifically, diffraction-limited resolution is achieved throughout the sample, not just within focal volume of the illuminating field. Simulated and experimental verifications of this effect are presented. ISAM's computational focusing obviates the trade-off between lateral resolution and depth-of-focus seen in traditional OCT.
{"title":"Interferometric Synthetic Aperture Microscopy: Physics-Based Image Reconstruction from Optical Coherence Tomography Data","authors":"B. Davis, T. Ralston, D. Marks, S. Boppart, P. Carney","doi":"10.1109/ICIP.2007.4379975","DOIUrl":"https://doi.org/10.1109/ICIP.2007.4379975","url":null,"abstract":"Optical coherence tomography (OCT) is an optical ranging technique analogous to radar - detection of back-scattered light produces a signal that is temporally localized at times-of-flight corresponding to the location of scatterers in the object. However the interferometric collection technique used in OCT allows, in principle, the coherent collection of data, i.e. amplitude and phase information can be extracted. Interferometric synthetic aperture microscopy (ISAM) adds phase-stable data collection to OCT instrumentation and employs physics-based processing analogous to that used in synthetic aperture radar (SAR). That is, the complex nature of the coherent data is exploited to give gains in image quality. Specifically, diffraction-limited resolution is achieved throughout the sample, not just within focal volume of the illuminating field. Simulated and experimental verifications of this effect are presented. ISAM's computational focusing obviates the trade-off between lateral resolution and depth-of-focus seen in traditional OCT.","PeriodicalId":131177,"journal":{"name":"2007 IEEE International Conference on Image Processing","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134064034","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2007-11-12DOI: 10.1109/ICIP.2007.4379259
M. Raphan, Eero P. Simoncelli
Image denoising methods are often based on estimators chosen to minimize mean squared error (MSE) within the sub-bands of a multi-scale decomposition. But this does not guarantee optimal MSE performance in the image domain, unless the decomposition is orthonormal. We prove that despite this suboptimality, the expected image-domain MSE resulting from a representation that is made redundant through spatial replication of basis functions (e.g., cycle-spinning) is less than or equal to that resulting from the original non-redundant representation. We also develop an extension of Stein's unbiased risk estimator (SURE) that allows minimization of the image-domain MSE for estimators that operate on subbands of a redundant decomposition. We implement an example, jointly optimizing the parameters of scalar estimators applied to each subband of an overcomplete representation, and demonstrate substantial MSE improvement over the sub-optimal application of SURE within individual subbands.
{"title":"Optimal Denoising in Redundant Bases","authors":"M. Raphan, Eero P. Simoncelli","doi":"10.1109/ICIP.2007.4379259","DOIUrl":"https://doi.org/10.1109/ICIP.2007.4379259","url":null,"abstract":"Image denoising methods are often based on estimators chosen to minimize mean squared error (MSE) within the sub-bands of a multi-scale decomposition. But this does not guarantee optimal MSE performance in the image domain, unless the decomposition is orthonormal. We prove that despite this suboptimality, the expected image-domain MSE resulting from a representation that is made redundant through spatial replication of basis functions (e.g., cycle-spinning) is less than or equal to that resulting from the original non-redundant representation. We also develop an extension of Stein's unbiased risk estimator (SURE) that allows minimization of the image-domain MSE for estimators that operate on subbands of a redundant decomposition. We implement an example, jointly optimizing the parameters of scalar estimators applied to each subband of an overcomplete representation, and demonstrate substantial MSE improvement over the sub-optimal application of SURE within individual subbands.","PeriodicalId":131177,"journal":{"name":"2007 IEEE International Conference on Image Processing","volume":"1119 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134404512","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2007-11-12DOI: 10.1109/ICIP.2007.4379985
L. H. Fonteles, M. Antonini
LVQ is a simple but powerful tool for vector quantization and can be viewed as a vector generalization of uniform scalar quantization. Like VQ, LVQ is able to take into account spatial dependencies between adjacent pixels as well as to take advantage of the n-dimensional space filling gain. However, the design of a lattice vector quantizer is not trivial particularly when one wants to use vectors with high dimensions. Indeed, using high dimensions involves lattice codebooks with a huge population that makes indexing difficult. On the other hand, in the framework of wavelet transform, a bit allocation across the subbands must be done in an optimal way. The use of VQ and the lack of non asymptotical distortion-rate models for this kind of quantizers make this operation difficult. In this work we focus on the problem of efficient indexing and optimal bit allocation and propose efficient solutions.
{"title":"High Dimension Lattice Vector Quantizer Design for Generalized Gaussian Distributions","authors":"L. H. Fonteles, M. Antonini","doi":"10.1109/ICIP.2007.4379985","DOIUrl":"https://doi.org/10.1109/ICIP.2007.4379985","url":null,"abstract":"LVQ is a simple but powerful tool for vector quantization and can be viewed as a vector generalization of uniform scalar quantization. Like VQ, LVQ is able to take into account spatial dependencies between adjacent pixels as well as to take advantage of the n-dimensional space filling gain. However, the design of a lattice vector quantizer is not trivial particularly when one wants to use vectors with high dimensions. Indeed, using high dimensions involves lattice codebooks with a huge population that makes indexing difficult. On the other hand, in the framework of wavelet transform, a bit allocation across the subbands must be done in an optimal way. The use of VQ and the lack of non asymptotical distortion-rate models for this kind of quantizers make this operation difficult. In this work we focus on the problem of efficient indexing and optimal bit allocation and propose efficient solutions.","PeriodicalId":131177,"journal":{"name":"2007 IEEE International Conference on Image Processing","volume":"103 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134539181","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：