SBIR Phase II: Extraction of Object Semantics from Paper Blueprints

*FURICK 9521774这个小型企业创新研究第二阶段项目将创建算法和软件，将传统的纸质蓝图转换为电子形式。目前，还没有将纸质蓝图无缝转换为电子形式的方法。现有的自动转换方法往往容易出错，并且无法捕获图纸的重要语义内容。现在，大多数向电子表格的转移都是通过手动将纸质蓝图绘制成电子表格进行的。这既昂贵又不可靠。因此，对现有结构的检查和评估是有限的，特别是在发生危机时或发生故障后。在这个第二阶段的计划中，将创建新的算法来提高栅格到矢量转换的速度和精度。该计划的第二个目标将是以基于知识的对象格式捕获这种几何图形和文本。总体研究目标是创建可供非计算机专家的执业工程师使用的框架、方法和软件。这一第二阶段计划将允许对现有纸质蓝图进行基于现代计算机的分析。开发的软件将在帮助地震分析、故障预测以及用于日常翻译和存储档案图纸方面具有重要的商业潜力。

*** Furick 9521774 This Small Business Innovation Research Phase II project will create algorithms and software to convert legacy paper blueprints into electronic forms. At present, there is no seamless way to transfer paper blueprints into electronic form. Existing automatic transfer methods often are error-prone and fail to capture important semantic content of the drawings. Most transfers to electronic form now occur by manually tracing paper blueprints into electronic form. This is expensive and unreliable. As a result, examination and assessment of existing structures is limited, especially in time of crisis or after a failure has occurred. During this Phase II program, new algorithms will be created to improve the speed and accuracy of raster to vector conversions. A second goal of the program will be to capture this geometry and text in knowledge-based object formats. The overall research objective is to create frameworks, methods and software that can be used by a practicing engineer who is not a computer expert. This Phase II program will permit modern computer based analysis of existing paper blueprints. The software developed will have important commercial potential in helping earthquake analysis, failure predictions, as well as use in everyday translation and storage of archival drawings.