{"title":"Parallel image classification on the HIVE","authors":"M. Smit, J. Garegnani, M. Bechdol, S. Chettri","doi":"10.1109/AIPRW.2000.953601","DOIUrl":"https://doi.org/10.1109/AIPRW.2000.953601","url":null,"abstract":"Remotely sensed imagery represents a growing source of information to many practical applications. Technologies to rapidly process imagery data into useful information products has not kept pace with the rapidly growing volume and complexity of imagery data increasingly available from Government and commercial sources. Significant processing speed improvements have been achieved by implementation of classification methods on the highly-parallel integrated virtual environment (HIVE) - a Beowulf class system using parallel virtual machine software. This paper discusses our parallel processing architecture and how three different classification algorithms performed in this computing environment. Also discussed are conclusions and recommendations for future work to apply these techniques to more complex data and further improve the processing speeds.","PeriodicalId":312930,"journal":{"name":"Proceedings 29th Applied Imagery Pattern Recognition Workshop","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126085165","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}
{"title":"Image retrieval with relevance feedback","authors":"Li Fang, Y. Ang","doi":"10.1109/AIPRW.2000.953608","DOIUrl":"https://doi.org/10.1109/AIPRW.2000.953608","url":null,"abstract":"The proposed system for image retrieval using multidimensional features (IRMF) characterises and matches image content in a high dimensional feature space of colour, texture and shape dimensions. By including the entire pyramid of low-, medium-, and high-level primitives, the semantics of image content at different feature levels can be represented and extracted efficiently for image retrieval. This provides accurate query formulation and improves the accuracy in the search results. By co-jointly matching image features in a multidimensional space rather than in separate independent feature spaces, the precision in image retrieval is improved from more than 50% to up to 90% for the top 10 most similar images retrieved. The impact of the information of the image's background has been mentioned in a very few published papers. Our experiments show that the efficient extraction of background information can improve the precision of image retrieval. To speed up the retrieval process, we also propose interactive relevance feedback to let the user participate in the process. The system is implemented for Internet Web access.","PeriodicalId":312930,"journal":{"name":"Proceedings 29th Applied Imagery Pattern Recognition Workshop","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122950165","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}
{"title":"High accuracy registration and targeting","authors":"Brian Lofy","doi":"10.1109/AIPRW.2000.953630","DOIUrl":"https://doi.org/10.1109/AIPRW.2000.953630","url":null,"abstract":"Advanced programs at BAE SYSTEMS-Mission Solutions has implemented high accuracy registration and targeting (HART) unit, which is a system for automated, multi-sensor, multi-date, image-to-image registration. HART combines N-image registration with a rigorous photogrammetric adjustment algorithm to achieve an automatic and precise geopositioning of imagery. HART is comprised of three top-level functions: the edge extraction, edge matching, and photogrammetric adjustment functions. The edge extractor detects sensor-invariant edges in an area-of-interest (AOI) at three levels of resolution. The multiresolution edge matcher correlates those edges within the AOI. The photogrammetric adjustment function updates the sensor support data for all the imagery over the AOI. We describe applications of HART to automatically register synthetic aperture radar (SAR), infrared, and electrooptical images within 2-pixel accuracy for use in the areas of mapping, targeting and data fusion.","PeriodicalId":312930,"journal":{"name":"Proceedings 29th Applied Imagery Pattern Recognition Workshop","volume":"153 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115383618","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}
{"title":"Moving object tracking in video","authors":"Yiwei Wang, J. Doherty, R. E. V. Dyck","doi":"10.1109/AIPRW.2000.953609","DOIUrl":"https://doi.org/10.1109/AIPRW.2000.953609","url":null,"abstract":"The advance of technology makes video acquisition devices better and less costly, thereby increasing the number of applications that can effectively utilize digital video. Compared to still images, video sequences provide more information about how objects and scenarios change over time. However, video needs more space for storage and wider bandwidth for transmission. Hence is raised the topic of video compression. The MPEG 4 compression standard suggests the usage of object planes. If the object planes are segmented correctly and the motion parameters are derived for each object plane accordingly, a better compression ratio can be expected. Therefore, to take full advantage of the MPEG 4 standard, algorithms for tracking objects are needed. So, we propose an algorithm to track moving objects in video sequences. The algorithm first separates the moving objects from the background in each frame. Then, four sets of variables are computed based on the positions, the sizes, the grayscale distributions and the presence of textures of the objects. A rule-based method is developed to track the objects between frames, based on the values of the variables. Preliminary experimental results show that the algorithm performs well. The tests also show that the algorithm obtains success in indicating new tracks, ceased tracks and possible collisions.","PeriodicalId":312930,"journal":{"name":"Proceedings 29th Applied Imagery Pattern Recognition Workshop","volume":"212 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123053224","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}
J. Mumbru, G. Panotopoulos, D. Psaltis, X. An, Gan Zhou, F. Mok
{"title":"Optically reconfigurable gate array","authors":"J. Mumbru, G. Panotopoulos, D. Psaltis, X. An, Gan Zhou, F. Mok","doi":"10.1109/AIPRW.2000.953607","DOIUrl":"https://doi.org/10.1109/AIPRW.2000.953607","url":null,"abstract":"Summary form only given, as follows. Reconfigurable processors, like the Field Programmable Gate Arrays (FPGAs), open new computational paradigms where the processor is able to tailor its internal structure to better implement a given application. A typical FPGA consists of an array of configurable logic blocks and a mesh of interconnections fully programmable by the user to perform a given application. By just changing its internal connectivity, the FPGA can implement a totally different new function. However in most of the applications, the FPGA is configured only once and used as coprocessor to carry out some highly complex or time-consuming computation. The reason for such limitation is the small communication bandwidth between the FPGA chip and the external memory, usually ROM, where the configuration data is stored. The Optically Programmable Gate Array (OPGA), an enhanced version of a conventional. FPGA, can overcome this problem. The OPGA utilizes a holographic memory accessed by an array of VCSELs to program its logic. The on-chip logic has been complemented with an array of photodetectors to detect the configuration template recorded in the memory. Combining spatial and shift multiplexing to store the configuration pages in the memory, the OPGA module is very compact and has extremely short configuration time allowing for dynamic reconfiguration. The reconfiguration capability of the OPGA can be applied to solve more efficiently problems in pattern recognition and database searches.","PeriodicalId":312930,"journal":{"name":"Proceedings 29th Applied Imagery Pattern Recognition Workshop","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131567436","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}
{"title":"A constant probability of detection model for image quantization","authors":"R. Bonneau","doi":"10.1109/AIPRW.2000.953612","DOIUrl":"https://doi.org/10.1109/AIPRW.2000.953612","url":null,"abstract":"A large number of remote sensing applications require that images be transmitted across a low bandwidth communication link for human analysis and automatic evaluation. To accommodate the low bandwidth link, compression is often used to reduce the amount of data transmitted. A central part of this compression process is quantization which decreases the entropy in the compressed imagery. While quantization decreases the amount of data it adds nonlinear distortion to the image. This nonlinear quantization noise can severely impair the ability of the analyst and automatic algorithm to identify features of interest in the decompressed data. We develop a wavelet Markov model based approach to detection before quantization to preserve features of interest to the analyst or automatic algorithm.","PeriodicalId":312930,"journal":{"name":"Proceedings 29th Applied Imagery Pattern Recognition Workshop","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124010787","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}
A. Deans, Hong Liu, W. Jung, Dechun Li, Xiaoguang Song, L. Fajardo
{"title":"Digital fluorescence technique for biomedical applications","authors":"A. Deans, Hong Liu, W. Jung, Dechun Li, Xiaoguang Song, L. Fajardo","doi":"10.1109/AIPRW.2000.953619","DOIUrl":"https://doi.org/10.1109/AIPRW.2000.953619","url":null,"abstract":"Summary form only given. We are investigating an optoelectronic imaging technique for localizing fluorescence probes in biologic specimens and experimental animals and analyzing the uptake, localization, and significance of these agents in biological structures. A prototype, digital optical imaging system consisting of a white light illuminator with an optical fiber light guide, a custom designed wavelength-selective filter module, and a highly sensitive, CCD imager was developed in our laboratory. For acquiring digital fluorescence images, the fluorescent probe is excited by the external light source with a selected wavelength. The fluorescent light emitted is then guided to the photosensitive surface of the detector through a dichroic mirror and a series of optical filters corresponding to the emission wavelength of the optical probe. The electronic output of the detector is then routed to a computer for analysis, storage and display. The system offers 14 bit digitization and the dynamic range is larger than 12 bits. Experiments and corresponding quantitative analysis demonstrate a high sensitivity for our prototype. The digital fluorescent imaging device was built with state-of-the-art, yet currently available optoelectronic technologies. It can be applied to cellular, molecular, histologic, embryologic and pharmacologic and other biomedical applications.","PeriodicalId":312930,"journal":{"name":"Proceedings 29th Applied Imagery Pattern Recognition Workshop","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130929639","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}
{"title":"An approach to image classification in ultrasound","authors":"M. Loew, Rashidus S. Mia, Zhenyu Guo","doi":"10.1109/AIPRW.2000.953625","DOIUrl":"https://doi.org/10.1109/AIPRW.2000.953625","url":null,"abstract":"Ultrasound is an attractive diagnostic medical imaging modality because of its safety, speed, relative low cost, and versatility. It is used successfully in imaging of shape, structure, and size in two and three dimensions; in the measurement of blood flow, and in quantitative characterization of tissue. The paper examines some of the models and processing techniques that make those applications possible. For example, quantitative ultrasound tissue characterization requires the estimation of parameters from the observed ultrasound radio-frequency signal. First, those parameters must be identified, second, algorithms used to extract those parameters from the observed signal must be developed; third, the classifier must be chosen; and finally, those extracted parameters are used to train the classifier, which is then tested to estimate classification performance. The methods are applied to the imaging of several kinds of liver tissue.","PeriodicalId":312930,"journal":{"name":"Proceedings 29th Applied Imagery Pattern Recognition Workshop","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128788725","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}
{"title":"Gradient-oriented profiles for unsupervised boundary classification","authors":"R. Tamburo, G. Stetten","doi":"10.1109/AIPRW.2000.953627","DOIUrl":"https://doi.org/10.1109/AIPRW.2000.953627","url":null,"abstract":"We present a method for unsupervised boundary classification by producing and analyzing intensity profiles. Each profile is created by sampling an ellipsoidal neighborhood of voxels oriented along the image gradient. The profile is analyzed via nonlinear optimization to find the best fitting cumulative Gaussian. The parameters of the cumulative Gaussian parameterize the boundary directly yielding: (1) extrapolated intensity values for voxels located far inside and outside of the boundary; and (2) estimates the boundary location and boundary width. For these parameters, intrinsic measures of confidence are established to eliminate low-confidence parameter estimates. Neighborhoods overlap considerably, yielding sufficient high-confidence estimates for a thorough survey of the boundary. Gradient oriented profiles are demonstrated on artificially generated 3D test data and proved to accurately parameterize and classify the boundary.","PeriodicalId":312930,"journal":{"name":"Proceedings 29th Applied Imagery Pattern Recognition Workshop","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114081181","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}
{"title":"Hierarchical, multi-resolution models for object recognition: applications to mammographic computer-aided diagnosis","authors":"P. Sajda, C. Spence, L. Parra, R. Nishikawa","doi":"10.1109/AIPRW.2000.953620","DOIUrl":"https://doi.org/10.1109/AIPRW.2000.953620","url":null,"abstract":"A fundamental problem in image analysis is the integration of information across scale to detect and classify objects. We have developed, within a machine learning framework, two classes of multiresolution models for integrating scale information for object detection and classification-a discriminative model called the hierarchical pyramid neural network and a generative model called a hierarchical image probability model. Using receiver operating characteristic analysis, we show that these models can significantly reduce the false positive rates for a well-established computer-aided diagnosis system.","PeriodicalId":312930,"journal":{"name":"Proceedings 29th Applied Imagery Pattern Recognition Workshop","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2000-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126543166","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}