Just to give you an idea, I listed a few examples. There are plenty of features to expand this model. In this article a very basic table model is described. This is done by using simple intersection criteria. The second step iterates through all generated lines and splits the contained words to columns. Because of noise elements (like OCR errors), Hugh transformations work better when restricted to small areas. That’s the reason why we don’t do global line segmentation. Wherever words have overlapping projection on the Y-axis, they are combined to a line element. This clustering does a so called Hugh-transformation. In the first step, the table’s lines are clustered from the selected word elements. The extraction process is implemented in two simple steps. Each column request provides the relative width referring to the table. This table request contains column requests. In order to represent our knowledge about the table, we create a table request. That is because of the special character of the cluster algorithm, which clusters words to lines. At this point we will not generate lines. After OCR process is done, we generate an instance of that model by converting the MODI objects. We don’t use the MODI Object model this time, because we need the line elements which are not provided in the MODI model. This is a hierarchy of four layout element classes: Documents, pages, lines, words. The Document Modelįor the application, we design a simple document model. I want to draw your attention to the underlying object model. The implementation neither includes special tricks nor does it provide breaking new design patterns. Of course, you may use the already customized table. To customize your table request, choose Add Columns and resize them by dragging the column headers.By default, the table request will contain only one single column. Select a table you want to extract by using the red selection area.Press the OCR button to get plain document text.The next steps guide you through the whole process of table extraction:
This is a semi-transparent tool window to customize your personal table requests. The new feature is the 'Table Capture Frame'. This application can be seen as an expanded version to the MODI example from Document Processing Part I. To show RDE technology in a simple way, I created the TableExtractor. After the process, you have one single information model for every document - a small but very important difference. With RDE you create a unique table pattern for all documents and let the machine create corresponding results. But since OCR cannot ‘know’ that all documents contain the same table layout, you will get (worst case) three different table formats. If you are not willing to type in every character manually, you can scan the documents and perform an OCR analysis. In our example, you have 3 paper documents with the very same table layout (your phone bills for example). Here is an example: Let’s say, you want to export table data to an Excel file. So why do we need extra table extraction? The Problem They also provide higher layout structures like lists and tables. Today’s OCR systems are no longer restricted to read floating text passages.
OCR is a powerful and popular technique to read paper based documents. In this second part of the overview, the subject is Request Driven Extraction (RDE) as the next step beyond plain OCR analysis. Document processing is used since decades in the financial and insurance industry.
0 kommentar(er)
