CAJ | 학술논문

为有效提取PDF(portable document format)文件中的文字、图片、图形信息,提出了包含文件预处理、显示预处理、功能扩展、显示4个单元的PDF阅读器的实现模型.基于PDF文件结构特点,提出了忽略次要信息定位关键位置的解析思路.在此基础上,针对FlatcDecode、DCTDecode和CCITTFFaxDecode这3种过滤器处理的数据流,给出了详细的解决方案,然后对PDF页面内容进行两次解析,设计相应的文字图形等数据结构保存结果,最后对数据利用和功能扩展进行了讨论.通过实验结果表明,该模型能较好地实现PDF信息提取和显示,有利于PDF在中文信息处理领域中的进一步开发利用.
위유효제취PDF(portable document format)문건중적문자、도편、도형신식,제출료포함문건예처리、현시예처리、공능확전、현시4개단원적PDF열독기적실현모형.기우PDF문건결구특점,제출료홀략차요신식정위관건위치적해석사로.재차기출상,침대FlatcDecode、DCTDecode화CCITTFFaxDecode저3충과려기처리적수거류,급출료상세적해결방안,연후대PDF혈면내용진행량차해석,설계상응적문자도형등수거결구보존결과,최후대수거이용화공능확전진행료토론.통과실험결과표명,해모형능교호지실현PDF신식제취화현시,유리우PDF재중문신식처리영역중적진일보개발이용.
To extract the text, images and graphical information from PDF file validly, an implementation model including four units (file pretreatment, display pretreatment, function extension and display) is raised. Based on the structure of PDF file, a solution of ignoring secondary message and positioning key information is put forward. On this basis, a solution to the data stream processed by FlateDecode, DCTDecode and CCITTFaxDecode filters is presented. After analyzed PDF pages twice, corresponding data structure of text and graphical are designed to record the results. At last the data utilization and function extension are discussed. The model can implement the extraction and display of information in PDF file well by experimental comparison, and it will benefit the further development of PDF in the field of Chinese information processing.