CN101334814A - Automatic scanning and reading system and reading method - Google Patents

Automatic scanning and reading system and reading method Download PDF

Info

Publication number
CN101334814A
CN101334814A CN 200810105224 CN200810105224A CN101334814A CN 101334814 A CN101334814 A CN 101334814A CN 200810105224 CN200810105224 CN 200810105224 CN 200810105224 A CN200810105224 A CN 200810105224A CN 101334814 A CN101334814 A CN 101334814A
Authority
CN
Grant status
Application
Patent type
Prior art keywords
volume
information
system
test
module
Prior art date
Application number
CN 200810105224
Other languages
Chinese (zh)
Inventor
仲莉恩
吴克河
周登文
斌 廖
柳长安
程文刚
俊 陆
陈圣俭
Original Assignee
华北电力大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date

Links

Abstract

The invention discloses an automatic scanning examination paper marking system and a corresponding method, belonging to the image processing and mode identifying technique field. The system comprises a paper defining module, an answer sheet manufacturing tool based on VBA, a file scanner, a database server, an answer sheet information automatic identifying module, an answer sheet identified-result revising module and an automatic marking summarizing module. The whole structure of the automatic paper marking system appears as a star and the database server is the center of the system and other modules and apparatuses are respectively connected with the database server by network means and used for information transmission and network sharing. One automatic scanning examination paper marking method includes the following steps: paper defining, paper manufacturing based on VBA, examination, subjective question marking, answer sheet scanning, answer sheet information identifying, identified-result revising and automatic marking and summarizing. The advantage is that the invention can mark the paper including both subjective questions and objective questions; the requirements to the operation technique and apparatus are low; the requirements to the paper-quality of the answer sheet, pens and printing are low and the checking is convenient.

Description

一种自动化的扫描阅巻系统及阅巻方法技术领域本发明属于图像处理和模式识别技术领域,特别是提供了一种自动化的扫描阅巻系统,利用数字图像处理和模式识别技术对答巻信息进行自动识别的系统和阅巻方法, 尤其涉及一种用于包含客观题和主观题试巻的自动阅巻方法。 An automated scanning and reading system reading Volume Volume TECHNICAL FIELD The present invention belongs to the technical field of image processing and pattern recognition, in particular to provide an automated scanning Volume reading system, digital image processing and pattern recognition information Volume fluent Volume reading system and method for automatic identification, in particular objective questions and subjective questions automatic test Volume Volume relates to a method for reading comprising. 背景技术考试是评估学生知识掌握程度的一种最常用、最普遍的手段,通过各种规模的考试有利于教师教学工作的推进、学生学习成绩的提高和整个教育水平的提升。 BACKGROUND exam is evaluated for one to grasp the extent of the students' knowledge of the most common and most popular resort, through the examination of all sizes to help promote the teaching of teachers, improve and enhance the overall educational level of student achievement. 客观题可以检测学生的判断能力,而主观题则可以反应出学生的逻辑推理、总结概括的能力。 Objective questions can detect students 'ability to judge, and subjective questions can reflect the students' logical reasoning, the ability to summarize general. 因此, 一份既含有客观题又含有主观题的试巻具有相当的测试价值。 Therefore, one containing both objective and subjective questions and tried Volume containing test questions have considerable value. 实现自动化扫描试巻的评阅是一项涉及多方面技术的问题,目前国内外在这方面都进行了一些研究和实践。 Automated scan test Volume reviewers is a multifaceted problem involving technology, in this respect at home and abroad have carried out some research and practice. 应用最多的是光学标记阅读机(OMR, Optical Mark Reader), 它基于光电转换的功能原理,以快速、准确的性能使得其在国内外标准化考试的阅巻、 评分等信息处理系统中得到普遍应用;但与此同时,它还存在着仅能评阅客观题、对答题卡纸张质量要求高、对信息卡印刷精度要求高、对填涂区灰度的浓度、颜色要求高以及需要专门的光标阅读机器等问题。 The most widely used optical mark reader (OMR, Optical Mark Reader), which is based on the principle of photoelectric conversion features to fast and accurate performance makes it widely used in domestic and international standardized tests of reading Volume, ratings, and information processing system ; At the same time, it exists only reviewers objective questions, the answer to the high quality paper card requires high precision printing of information cards, Tiantu concentration area gradation, high color requirements and require special OMR machinery and other issues. 另外一种在大型选拔考试(例如,高考)中广泛使用的是网上阅巻系统,它采用试巻和答巻分离的方式,主观题由不同阅巻教师通过网络在计算机上对考生答巻的电子图像分别进行评分,最终由计算机系统自动进行核分和成绩校验的一种阅巻方式。 Also widely used in a large selection examination (for example, college entrance examination) are online reading Volume system, which uses tried and Volume Volume A separate ways, subjective questions by different teachers by reading Volume Volume A network of candidates on a computer electronic images were scored, eventually one kind of nucleating points and reading Volume way check results are automatically determined by the computer system. 网上阅巻有着以下优点:它采用一巻多评制(主观题有多个老师进行评阅)和误差控制机制,有助于阅巻教师更好地把握评分标准、 控制误差、提高阅巻质量;但同时存在着对阅巻环境、扫描设备、答题纸质等要求高, 此外庞大的阅巻成本也使其难以在普通考试中推广应用。 Volume online reading has the following advantages: It uses a multi-evaluation system Volume (subjective question there were more teachers reviewers) and error control mechanisms to help teachers better understand reading Volume scoring criteria, control errors, improve the quality of reading Volume; but at the same time there is a high reading on the Volume environment, scanning equipment, the answer paper and other requirements, in addition to the huge costs make it difficult to read Volume popularized in the ordinary exam. 我国基础教育阶段的考试有着频繁性的特点。 China's basic education stage of the exam has the characteristics of frequent nature. 为了能及时地检测学生对知识的掌握程度,促进学生知识水平的提高,除了常规的期中、期末考试外,还增加了月考、 小测试等各种科目、各种规模的考试。 In order to timely detect students' mastery of knowledge, promote the improvement of the level of knowledge of students, in addition to conventional midterm, final exam, also increased the monthly tests, quizzes, and other subjects, the examination of all sizes. 与频繁的考试相对应的是,基础教育阶段的考试迫切需要快速、准确、小成本地阅巻。 And frequent examinations Correspondingly, basic education stage of examination urgent need for rapid, accurate and cost less to read Volume. 教师要在较短的时间内评阅试巻,时间之紧、 工作量之大容易使得各个环节不可避免地出现人工误差,而且试巻登分、核分、保管等各项工作需要耗费人力、物力以及场地等资源。 Teachers should try reviewers Volume in a short period of time, the time is tight, it is easy to make a large amount of work in all aspects of human error inevitably arise, and board test points Volume, nuclear sub, storage and other work takes manpower and material resources and resources venues. 因此,为了能够满足我国基础教育阶段考试的需要, 一种自动的扫描阅巻系统的出现就显得尤为重要了。 Therefore, in order to meet the needs of China's basic education stage of examination, appears an automated scanning system of reading Volume is particularly important. 发明内容本发明的目的在于提供一种自动化的扫描阅巻系统及阅巻方法,有效地解决了传统人工阅巻存在的时间紧、任务重、査找纸质试巻工作量大、登分和核分保存困难以及耗时耗力等问题。 SUMMARY OF THE INVENTION An object of the present invention to provide an automated scanning and reading system reading Volume Volume method, effectively solve the problems of the conventional artificial Volume read time, heavy task, Volume lookup workload test paper, board and nuclear division save difficult and time-consuming sub-issues. 提高了教师的阅巻质量,减轻了教师的劳动负担,有效地减少人工处理及答巻周转环节中的差错,省时省力,而且试巻图像可存档,为提高阅巻质量' 增添了一层保障,最大限度地保障评巻公平、公正,并且对电子化教学管理有重大的意义和影响。 Volume improve the quality of reading teachers, reducing the labor burden on teachers, effectively reducing manual handling and answer session Volume turnover errors, saving time and effort, and test Volume image can be archived, in order to improve the quality of reading Volume 'adds a layer security, maximum protection for comment Volume fair, just, and of great significance and impact of e-learning management. 本发明采用的技术方案是:采用试巻和答巻分离的方式,设计出一种包含主观题和客观题的答巻的自动识别信息系统和方法,不仅使得教师可以方便快捷地制作答巻, 而且能够高效地实现答巻分数等相关信息的识别。 Aspect of the present invention is that: Volume A and Volume test using isolated manner, the automatic identification information design system and method comprising a Volume A subjective questions and objective questions, so that the teachers can not only easily and quickly produced Volume A, a relevant identification information and Volume fraction as can be achieved efficiently. 本发明的系统由以下部分组成:试巻定义模块、基于VBA的答巻制作工具、文档扫描仪、数据库服务器、答巻信息自动识别模块、答巻识别结果校正模块、自动评分汇总模块。 The system of the present invention consists of the following components: Volume test definition module, Volume Builder based on A, document scanner, the database server VBA, Volume Information recognition module A, Volume A recognition result correction module, automatic scoring summary module. 整个自动阅巻系统是一个星形结构,数据库服务器是系统的中心,其他各个模块及设备都单独以网络的形式与其相连接,用于信息传递和网络共享。 Volume entire automatic reading system is a star-shaped configuration, the database server is the center of the system, and each of the other modules in the form of separate devices connected to its network, and for transmitting network sharing information. 试巻定义模块是整个系统的第一个组件,安装在用户的计算机上,用于试巻基本信息、结构信息以及答案与评分标准的定义,并把设置结果存储到数据库服务器中,为答巻制作提供信息;基于VBA的答巻制作工具是系统的第二个组件,安装在用户的计算机上,用于生成答巻的电子模板,其输出结果转化为标准答巻图像和考试实际使用的答巻;系统的第三个组件是文档扫描仪,安装在扫描人员的设备间,用来接收大量的考生答巻,对其电子化并自动转存到数据库服务器中;答巻信息自动识别模块是系统的第四个组件, 安装在用户的计算机上,它从数据库服务器中获取考生答巻,之后进行自动识别,最后将识别结果返回给数据库服务器;系统的第五个组件是答巻信息校正模块,安装在用户的计算机上,从数据库服务器获取自动阅巻的结果信息并显示,供 Volume test definition module is the first component of the system, installed on the user's computer, Volume test for basic information, configuration information, and answers the definition score, and the setting result is stored in the database server, to A. Volume production information; VBA-based builder Volume a is the second component of the system, installed on the user's computer, the electronic template for generating a Volume a, which is converted to the output image and the standard test a a Volume actually used Volume; the third component of the system is a document scanner, the scanning device mounted between the personnel, for receiving a large number of candidates Volume a, and its electronic autodump to the database server; Volume information recognition module a is the fourth component of the system, installed on the user's computer, which gets from the database server candidates Volume a, followed by automatic recognition, the final recognition result returned to the database server; the fifth component of the system information of the correction module is A. Volume installed on the user's computer, read Volume automatically acquired from the database server and displaying results information for 考务人员校正, 并将校正结果返回存储到数据库服务器上;自动评分汇总模块是系统的第六个组件, 安装在用户的计算机上,从数据库获取识别校正的结果数据,并根据试巻定义存储的信息,进行自动的评分与汇总。 Proctors correction, and the correction result stored back to the database server; automatic scoring module is a summary of the sixth component of the system, installed on the user's computer, acquires the identification result of the correction data from the database and stored in accordance with the test defined Volume information, automatically score and summary. 打印机是系统的可选组件,与自动评分汇总模块所在的计算机相连接,用于打印输出成绩等统计数据。 The printer is an optional component of the system, and a computer module located summary score automatically connected, and the like for printout performance statistics. 本发明的自动扫描阅巻方法包括以下几个步骤:a. 试巻定义,是指利用本系统的试巻定义模块,完成试巻结构的定义和答案与评分标准的设置。 Volume auto scan reading method of the present invention comprises the following steps:. A test defined Volume, Volume is the use of the test modules of the system is defined, the definition is completed, and the scoring of the answers provided Volume test structure. 步骤a还包括以下步骤:al.试巻结构定义,是命题教师利用本系统的试巻定义模块,定义试巻的结构。 A step further comprises the step of:. Al Volume test structure definition, the proposition teacher Volume test using the present system definition module defines Volume test structure. 首先,需要创建一份试巻的基本信息,包括满分值、考试年级、考试科目等信息;其次,创建大题,设置大题的主客观属性、大题数目和大题满分值;然后针对每道大题设置所包含的小题数目及各个小题的满分值,并把相应的信息存入数据库服务器中。 First, you need to create a test Volume of basic information, including out value, exam grades, test subjects and other information; secondly, to create a big issue, big issue set of objective and subjective attributes a large number of questions and a large value out of the question; then for each small number of big question title track and each set included a small value out of question, and the corresponding information stored in the database server. a2.答案与评分标准设置,是指利用本系统的试巻定义模块,在完成试巻结构的定义后,在系统中录入各个客观题的标准答案以及多选题的评分标准,而后试巻定义模块会将这些信息存入数据库服务器中。 a2. The answer is provided with score, refers to the use of this test system definition module Volume, Volume after completion of the test structure is defined, and the input standard answer a multiple-choice objective that each score in the system, then the test is defined Volume the information will be stored in the module database server. b. 基于VBA的答巻制作,是指利用本系统提供的基于VBA的试巻制作工具,根据已定义的试巻结构制作出答巻的电子模板,并把答巻模版存入数据库服务器中。 B. A Volume produced on the VBA, refers to the use of the test based on VBA Volume production tools, making the test system provided in accordance with a defined structure of Volume Volume A template of electron, and the Volume A is stored in the template database server. 己经制作好的电子模板主要用于以下两个方面: 一是生成标准答巻图像,二是作为考生实际考试使用的答巻。 Already making good electronic template is mainly used for the following two aspects: First generation A standard image Volume, Volume two is used as the candidates answer the actual exam. 在本发明中,标准答巻图像是指根据已制作好的答巻电子模板,利用Microsoft Office Document Image Writer的功能而直接生成的TIFF图像,从标准答巻图像识别出的坐标信息用于考生答巻图像的信息获取。 In the present invention, refers to the standard image Volume A. The electronic Volume A ready-made template, using Microsoft Office Document Image Writer functions TIFF images directly generated from the image recognition Volume A standard coordinate information of the candidates for A Volume image information acquisition. c. 考试,是指考生根据考试安排参与考试的过程。 c. examination, refer to the candidates based on examination arrangements involved in the process of examination. 考试之前,考务人员需要将考生的考号收集并存储到数据库服务器中。 Before the exam, candidates need to Proctors candidate number collected and stored in the database server. 考试过程中,考生需要书写并填涂答巻中的考生信息区域、将客观题的答题结果填涂在答巻的客观题区域以及将主观题的答题结果书写在答巻的主观题答题区域。 Examination process, candidates need to write and Tiantu A candidate information in the area Volume, the result will answer objective questions TianTu A Volume in the objective area and will answer questions the results of subjective questions to answer subjective questions in writing area A of Volume. d. 主观题评阅,是指阅巻人员对考生答巻中主观题评阅的过程,在评阅完成之后, 阅巻人员将主观题的得分填涂在考生答巻的主观题分数涂卡区中。 d. reviewers subjective question, refers to the person reading Volume Volume on the candidates in the process of reviewers subjective question and answer, after the completion of reviewers, read Volume will score the subjective question of TianTu subjective question Score Tuka District candidates answer the Volume. e. 答巻扫描,是指利用文档扫描仪,将主观题评阅过的考生答巻成批扫描成一个个TIFF格式的图像文件(通常,同一试室或班级的考生的答巻会被扫描成一个TIFF文件),所得的图像存入数据库服务器内,用于后续的考生答巻信息识别之用。 e. A Volume scanning, is the use of document scanners, the subjective questions of the candidates through reviewers Volume A batch scanned into a TIFF image file format (usually, the same test room or the candidate's class A Volume will be scanned a TIFF file), the resulting image is stored in the database server for subsequent candidates A. Volume informational purposes. f. 答巻信息识别,是指利用本系统的答巻信息识别模块,识别出标准答巻和考生答巻中标记位、考号、客观题和主观题等区域的信息,并将识别信息存入到数据库服务器中。 f. A Volume identification information, the present system is the use of Volume A module identification information, and identifies criteria A candidate Volume of Volume A of the information region flag, test number, title objective and subjective questions, etc., and the identification information stored into the database server. 步骤f还包括以下两个步骤:fl.标准答巻信息识别,是识别并记录标准答巻图像的各部分(包括标记位、考号、客观题区域、各个主观题答题区域、各个主观题涂卡区域)的坐标信息,之后答巻信息识别模块将这些坐标信息存储到数据库服务器中。 Step f further comprises the following two steps:. Fl Volume A standard identification information, it is to identify and record portions of Volume A standard image (including the flag, test number, objective questions regions, each region answer subjective questions, each coating subjective questions card area) coordinate information after the identification information of Volume a stores the coordinate information of the module to the database server. f2.考生答巻信息识别,是识别出考生答巻的填涂信息。 f2. Volume A candidate identification information, is identified in Volume A are marked on the candidate information. 这是在标准答巻图像信息识别完成的基础上进行的,首先将倾斜的考生答巻旋转并校正到与标准答巻图像统一的坐标系上,然后根据位置标准答巻图像中各部分的坐标获取各个识别项的位置信息,以识别出考生答巻的填涂信息。 This is the standard A based Volume-image information identification completion performed on the first inclined candidates A. Volume rotated and corrected to standard A on Volume image unified coordinate system, the coordinates Volume image of each part and then A. The position of the standard acquiring position information of each identification item, information to identify candidates Tiantu Volume of a. g. 答巻识别结果校正,是利用本系统的答巻识别结果校正模块,将错误的识别信息进行校正。 g. Volume A recognition result correction, using this system module A. Volume recognition result correction, error correction of the identification information. 为了提高系统的准确性,对考生答巻进行图像分割,分割成为考号区域图像、考生信息区域图像(如姓名、学校和班级等)、客观题区域图像以及各个主观题区域图像。 In order to improve the accuracy of the system, A Volume of candidates for image segmentation, segmentation becomes, objective questions, subjective questions and the various regions of the image area of ​​the image area of ​​the image candidate number, candidate information area of ​​an image (such as name, school and class, etc.). 根据数据库服务器中的考号信息,自动判断已识别考号的有效性(包括考号是否识别完全、考号是否存在、考号是否重复出现等情况);对于主观题,根据当前题目的满分值判断教师所判分数是否合理等。 The candidate number information in the database server automatically determines the validity of the identified candidate number (including the identification number whether a complete test, test number exists, whether the number of repeated examination, etc.); For subjective questions, based on the current value of the subject out the judge sentenced the teacher scores is reasonable. 针对各种不合理的情况,给出友好明确的提示。 For all kinds of unreasonable situation, giving a clear and friendly tips. 答巻识别结果校正模块提供了各个切割图像以及相应的答案识别(针对客观题)或评分(针对主观题)结果,阅巻人员可以据此实现校正,校正后的信息将存储到数据库服务器中。 A Volume recognition result correction module provides various cutting image and the corresponding answers recognition (for objective questions) or scores (for subjective question) results, read Volume personnel could then achieve correction information after the correction will be stored in the database server. h. 自动评分汇总,是利用本系统的自动评分汇总模块,根据自动识别和校正后的客观题答案与主观题评分识别结果,依据评分规则,自动评分汇总每份答巻的分数。 h. automatic scoring summary, this system is the use of automatic scoring summary module, based on objective answer questions after the automatic identification and correction of problems with subjective scoring recognition result, according to the scoring rules, automatic scoring Volume A summary of each score. 本发明的有益效果是:本发明的答巻信息自动识别方法能够高效地识别出答巻的各项信息,准确率达到99%以上,识别结果基本不需要校正,极大地提高了阅巻的效率,降低了阅巻过程的复杂度,减少了人工误差,提高了透明度,减轻了考务人员工作量,最大限度地实现了考试的公开公平性。 Advantageous effects of the invention are: A. Volume Information recognition method of the present invention can efficiently recognize the information of the Volume A, the accuracy rate of 99% or more, substantially no correction recognition result, which greatly improves the efficiency of reading Volume , reducing the complexity of reading Volume process, reducing human error, increase transparency and reduce the workload of the Proctors, to maximize the fairness of public examinations. 自动化扫描阅巻系统有良好的经济效益和社会效益-1. 自动阅巻系统,性能稳定、识别准确和实时性好,自动阅巻的阅巻速度在l分钟30页以上,准确率在99%以上。 Volume automated scanning reading system with good economic and social benefits -1 Volume automatic reading system, stability, recognition accuracy and real-time, automatically read the read Volume Volume l speed is above 30 minutes, the accuracy rate of 99% the above. 2. 硬件设备较少,技术要求低,只要会使用办公软件Word、高速扫描仪以及本发明中的软件工具即可。 2. The small hardware, low technical requirements, so long as the use of office software Word, in the high-speed scanner and software tools to the present invention. 3. 对答巻纸张质量要求低,使用普通的复印纸即可,靠图像的标记位定位而不需卡纸定位。 3. Low Volume fluent quality paper, plain paper copy can, by positioning the image of the flag without jams positioned. 4. 对答题用笔要求低,学生答题过程和阅巻人员主观题判分过程中可用铅笔、钢笔、圆珠笔等各种书写笔填涂,而不特定要求专用的2B铅笔,尊重了学生的答题习惯。 4. Low requirement pen to answer subjective questions students answer sub-sub-process procedures and personnel available in the reading Volume pencils, pens, ballpoint pens and other writing Tiantu, without the specific requirements of a dedicated 2B pencil, respect for the student's answer habit. 5. 对答巻印刷要求低,系统能够处理倾斜、倒置以及小变形等情况。 5. Low Volume fluent printing requirements, the system can handle is inclined, inverted, and small deformation situation. 6. 操作简单,阅巻人员只需批阅主观题,其余工作都用系统的各个模块实现,大大减轻相关人员工作量。 6. Simple, marking only need to read Volume subjective question, and the remaining work is implemented by the various modules of the system, greatly reducing the workload of the relevant personnel. 7. 阅巻之后的答巻能够返还给学生,有利于学生了解自己的答题情况,有针对性地加强自己的学习。 After the reading of 7. A Volume Volume can be returned to students, help students understand their own answer situation, targeted to strengthen their own learning. 8. 应用本系统,实现试巻文档的电子化,对于构建试题库、科学的分析试巻的信度和效度以及构建区域常模等有积极的作用。 8. The application of this system, the realization of electronic Volume test document for building the test database, the reliability and validity of the scientific analysis and test Volume norm like building region have a positive effect. 9. 从试巻定义到答巻制作,出题教师可以利用本系统方便快捷地实现。 9. Volume Volume produced from a sample is defined to A, the title teachers can use the system easily and quickly achieved. 10. 答巻图像存储在数据库中,校正时査阅方便,利于用户调整识别数据时有据可寻,提高阅巻的准确度和透明度。 10. A database, making it easy Now Volume correction image storage, adjustment of user identification data conducive evidence to be found, Volume improve reading accuracy and transparency. 11. 有效地降低了人工误差,免去了保存纸质试巻的劳累,大大节约了人力、物力,节省了各种费用的开销。 11. effectively reduces human error, eliminating the need for paper-saving test Volume of fatigue, significant savings in manpower, material resources, saving the cost of various fees. 附图说明图l是本发明的自动化扫描阅巻系统结构示意图。 FIG l is a schematic diagram Volume automated scanning reading system configuration of the present invention. 图2是本发明的阅巻方法的步骤示意图。 Volume 2 is a step of reading the method of the present invention. FIG. 图3是本发明的自动化扫描阅巻系统的试巻定义模块之功能模块图。 Volume 3 is a functional test of the module definition module of FIG Volume automated scanning reading system according to the present invention. 图4是本发明的阅巻方法所采用的答巻各要素说明图。 FIG 4 is an explanatory view of the elements A Volume Volume-through process of the present invention is employed. 图5是本发明的阅巻方法之标准答巻信息识别流程图。 FIG 5 is a flowchart illustrating information for identifying the standard method of reading Volume of Volume A of the present invention. 图6是本发明的阅巻方法之考生答巻信息识别流程图。 FIG 6 is a flowchart illustrating information for identifying the candidates of the method of the present invention read Volume Volume answer. 具体实施方式本发明用于含有主观题和客观题答巻的自动化扫描阅巻,主要原理是:采用试巻和答巻分离的方式,命题教师定义并制作考试答巻,待考试完毕后,教师评阅主观题, 而后考生的答巻通过扫描仪扫描成图像文件,利用图像处理和模式识别技术在计算机上对电子化了的考生答答巻进行评阅。 Specific embodiments of the present invention contains a subjective question and answer objective questions Volume Volume of automated scanning reading, the main principle is used: With test Volume and Volume A separate ways, teachers define and make propositions Volume exam answer until the examination is completed, teachers reviewers subjective question, and then answer the candidates Volume scanned into image files by scanner, electronic Volume of the candidates were dripping reviewers on a computer using image processing and pattern recognition technology. 其包括以下步骤:a. 试巻定义,是指利用本系统的试巻定义模块,完成试巻结构的定义和答案与评分的设置,并将相关设置信息存储到数据库服务器中,为答巻制作及自动评分汇总提供信息。 Comprising the steps of:. A test defined Volume, Volume again refers to the use of the system definition module, and answers to complete the definition with scores Volume test structure is provided, and an associated setting information stored in the database server, to A. Production Volume and automatic scoring summary information. al.试巻结构定义,是命题教师利用本系统的试巻定义模块,定义试巻的结构。 al. Volume test structure definition, the proposition teacher Volume test using the present system definition module structure definition of Volume test. 首先,需要创建一份试巻的基本信息,包括满分值、考试年级、考试科目等信息;其次,创建大题,设置大题的主客观属性、大题数目和大题满分值;然后针对每道大题设置所包含的小题数目及各个小题的满分值,并把相应的信息存入数据库服务器中。 First, you need to create a test Volume of basic information, including out value, exam grades, test subjects and other information; secondly, to create a big issue, big issue set of objective and subjective attributes a large number of questions and a large value out of the question; then for each small number of big question title track and each set included a small value out of question, and the corresponding information stored in the database server. a2.答案与评分标准设置,是指利用本系统的试巻定义模块,在完成试巻结构的定义后,在系统中录入各个客观题的标准答案以及多选题的评分标准,而后试巻定义模块会将这些信息存入数据库服务器中。 a2. The answer is provided with score, refers to the use of this test system definition module Volume, Volume after completion of the test structure is defined, and the input standard answer a multiple-choice objective that each score in the system, then the test is defined Volume the information will be stored in the module database server. b. 基于VBA的答巻制作,是指利用本系统提供的基于VBA的试巻制作工具,根据已定义的试巻结构制作出答巻的电子模板,并把答巻模版存入数据库服务器中。 B. A Volume produced on the VBA, refers to the use of the test based on VBA Volume production tools, making the test system provided in accordance with a defined structure of Volume Volume A template of electron, and the Volume A is stored in the template database server. 已经制作好的电子模板主要用于以下两个方面: 一是生成标准答巻(没有填图痕迹的答巻) 图像,二是作为考生的答巻。 Ready made electronic template is mainly used for the following two aspects: First generation A standard Volume (no mapping traces A Volume) image, and second as A Volume candidates. 步骤b具体还包括以下歩骤:bl.标记位的生成,命题教师利用答巻制作工具在答巻的左上、左下、右上自动生成三个圆形黑点,它们就是所谓的标记位。 DETAILED step b further comprises the step ho:. Bl-bit flag generated, the proposition teachers to use A A Volume Volume builder in the upper left, lower left, upper three circular black spots generated automatically, which is called a flag bit. 其中,左上与左下竖直对齐,左上与右上水平对齐。 Wherein the vertically aligned upper left and lower left, upper left and upper right horizontal alignment. 标记位一旦确定,就不再变动,它用来校正扫描后发生了倾斜的考生答巻图像。 Once the flag, no longer change, which had tilted candidates for Volume A scanned image is corrected. b2.考生信息的生成,考生信息包括考生的姓名、班级、学校、考号,根据需要时可增加A/B巻类型及缺考标记,这些信息已被预先制作为固定图像,当打开试巻模板后,无需出题教师干涉,试巻会自动加载该图像到固定位置。 b2. candidates generated, the candidate information comprises candidate's name, class, school, test number, can increase the A / B Volume Types absent when labeled according to needs, the information has been prepared in advance as a fixed image, when the opening Volume Test after the template without the topic and teacher interference, try Volume will automatically load the image into a fixed position. b3.考试名称的生成,考试名称包括考试时间地点和考试科目两部分,答巻制作工具会根据命题教师的输入自动加载到答巻文档的固定位置。 b3. The exam name generation, the examination time and place names include examination and test subjects in two parts, A Volume production tool is automatically loaded into a fixed position Volume A document based on input proposition teachers. b4.客观题的生成,命题教师根据试巻结构,确定客观题的数目,答巻制作工具根据客观题的数量自动选择合适的客观题图片模板(0〜100题)并自动加载到固定位置。 b4. to generate objective questions, according to test propositions teacher Volume structure, determine the number of objective questions, and the answers Volume authoring tool automatically selects the appropriate picture templates objective questions (0 to 100 questions) based on the number of objective questions and automatically loaded into a fixed position. b5.主观题的生成,命题教师根据本次考试的试巻情况,设置答巻所需的页数、 每页的栏数、每栏的题目数、每个题目的高度及分数,答巻制作工具根据这些设置自动生成相应的页、栏、主观题答题区。 b5. generation, proposition subjective question of teachers according to test Volume circumstances of the examination, set the desired Volume A number of pages, number of columns per page, each column title number, height and score for each topic, and the answers Volume Production tool automatically generates the corresponding page, column based on these settings, subjective questions answer District. 答巻制作工具自动判断每个主观题的分数,当分数小于等于9.5分时,增加一行涂卡分数的图片,当分数大于9.5分时,增加两行涂卡分数的图片。 A Volume authoring tool automatically determines the score of each subjective question, when the score is less than equal to 9.5 min, add a line Tuka fraction of the picture, when the score is greater than 9.5 min, add two lines to Tuka fraction of the picture. b6.标准答巻图像的生成,是指答巻制作工具调用Microsoft Office Document Image Writer的功能,将Word中设计完成的电子模板输出为TIFF格式的图像文件,并将该图像保存在数据库服务器上,用于获得标准答巻图像的位置信息。 b6. A standard Volume generated image refers to A. Volume production tool calls Microsoft Office Document Image Writer features that will complete the design of electronic Word template output as an image file in TIFF format, and save the image on the database server, a picture for the standard Volume of location information. b7.考生答巻的生成,以答巻电子模板为样板印刷后作为考生答巻。 b7. Candidates A generation of Volume, Volume answer to electronic template after template printing as candidates Volume answer. c. 考试,是指考生根据考试安排参与考试的过程。 c. examination, refer to the candidates based on examination arrangements involved in the process of examination. 考试过程中,考生需要书写并填涂答巻中的考生信息区域,并根据试巻题目将客观题的答题结果填涂在答巻的客观题区域,将主观题的答题结果书写在答巻的主观题答题区域。 Examination process, candidates need to write and candidate information Tiantu A Volume in the region and are marked on the objective questions based on the answer area A Volume Volume test results subject to objective questions, the results will answer subjective questions in writing of Volume A answer subjective questions area. d. 主观题评阅,是指阅巻人员对考生答巻中主观题评阅的过程,在评阅完成之后, 阅巻人员将主观题的得分填涂在考生答巻的主观题分数涂卡区中。 d. reviewers subjective question, refers to the person reading Volume Volume on the candidates in the process of reviewers subjective question and answer, after the completion of reviewers, read Volume will score the subjective question of TianTu subjective question Score Tuka District candidates answer the Volume. e. 答巻扫描,是指利用文档扫描仪,以一定的分辨率(150dpi左右),将主观题评阅过的考生答巻成批扫描成一个个TIFF格式的图像文件(通常,同一试室或班级的考生的答巻会被扫描成一个TIFF文件),所得的图像存入数据库服务器内,用于后续的考生答巻信息识别之用。 e. Volume A scan, using the document scanner means, a certain resolution (about 150dpi), the candidates had subjective question reviewers Volume Batch A scanned into a TIFF image file format (typically, the same test chamber or a class Volume candidates will be scanned into a TIFF file), the resulting image is stored in the database server for subsequent candidates A. Volume informational purposes. f. 答巻信息识别,是利用本系统的答巻信息识别模块,识别出标准答巻和考生答巻图像中的相关信息,并把识别信息存入到数据库服务器中。 F. A Volume identification information, the present system is the use of Volume A module identification information, and identifies criteria A Volume Volume A candidate image information, and stores the identification information to the database server. 步骤f含有以下步骤-fl.标准答巻信息识别,是识别并记录标准答巻图像的各部分(包括标记位、考号、客观题区域、各个主观题答题区域、各个主观题涂卡区域)的坐标信息,之后答巻信息识别模块将这些坐标信息存储到数据库服务器中。 Step f comprises the following steps -fl. Volume A standard identification information, is to identify and record portions of Volume A standard image (including the flag, test number, area objective questions, the answer subjective questions respective regions, each region subjective question Tuka) coordinate information after the identification information of Volume a stores the coordinate information of the module to the database server. 歩骤gl的还包括如下步骤:ni标准答巻图像读取与预处理,是指利用系统的答巻信息识别模块将每个答巻图像从标准答巻图像包TIFF文件包中分离出来,读入内存并转化成256个灰度级别BMP格式的图像,之后对BMP图像进行二值化。 Ho gl step further comprises the step of: ni Volume A standard image preprocessing read, the system refers to the use of Volume A module identification information of each image from the standard A A Volume Volume picture pack package separate TIFF files, read into memory and converted to 256 gray levels of image BMP format, BMP images after binarization. fl.2.标记位识别与定位,是指识别并记录标记位的坐标值。 fl.2. flag identification and location, it means to identify and record the coordinate value of the flag bit. 基于模板匹配方法, 识别答巻中的左上、左下和右上等三个位置的标记位坐标,并记录到数据库服务器中, 这些坐标信息用于旋转校正扫描过程中发生了倾斜的考生答巻,以使其与标准图像在同一个坐标系中。 Template matching method based on coordinates of the three flag bits identifying the location of Volume A of the upper left, upper right and lower left, etc., and records to the database server, the coordinate information for these candidates had tilted Volume A scanning rotation correction process to so that the standard image in the same coordinate system. H.3.考号信息坐标定位,是指识别并记录考号区域及每个具体考号的坐标。 H.3. Coordinate positioning candidate number information, and means to identify candidate number region coordinate and each particular candidate number recording. 根据标记位坐标与考生区域相对位置的信息,分别向X轴、Y轴投影,投影值与阈值对比, 从而识别出考号区域的边框坐标,之后在每个考号区域内中向Y轴投影获得10个数字的坐标,并按一定规则记录到数据库服务器中。 The information bits coordinates of the relative position of the candidate region numerals, respectively, compare the X axis, Y axis projection, the projection value with a threshold value, thereby identifying the border coordinates candidate number region, then the axis projected in the Y in each candidate number region obtain the coordinates of 10 digits, according to certain rules recorded in the database server. f1.4.客观题坐标定位,是识别出客观题区域及每道选择题的A、 B、 C、 D具体选项的坐标值。 F1.4. objective questions coordinate positioning, the objective is to identify problem areas and each multiple-choice questions, A, coordinate values ​​B, C, D specific options. 根据标记位与客观题区域相对位置的信息,分别向X轴、Y轴投影,并比将投影值与阈值对比来识别客观题区域的边框坐标,之后在其区域内分别向X轴、Y轴投影以识别各选项的横纵坐标信息,并按一定规则记录到数据库服务器中。 The information flag relative position of the objective area for the title, respectively, in the X-axis, Y-axis projection, and the ratio of the projection values ​​and the contrast threshold values ​​to identify the border coordinates objective questions region, after each of the X axis, Y axis in its region projection to identify horizontal and vertical coordinates for each option, according to certain rules recorded in the database server. f1.5.主观题坐标定位,是确定每个主观题的区域坐标。 f1.5. coordinate positioning subjective question, it is to determine the coordinates of each area subjective question. 先将BMP格式的答巻图像整体膨胀后投影,然后根据主观题答题区横竖栏线黑色像素较多的特点,分别向X轴、 Y轴投影,投影值与阈值对比从而获取边框线的坐标,并根据宽度高度等信息进行分组判断,从而得出属于同一道主观题答题区的边框线,由此获得主观题答题区坐标信息。 After the entire Volume A first expansion BMP format image projection, based on subjective question and answer areas, vertical bar line of black pixels more characteristics, respectively, to compare the X-axis, Y-axis projected with a threshold value to acquire the coordinates of the border line, the height and width information for the packet is determined to arrive at the border line belonging to the same answer subjective questions channel region, thereby obtaining the coordinate information area answer subjective questions. 以每个主观题答题区的下边线坐标为基准,继续使用投影的方法,得到涂卡分数区域的相对坐标信息,并按一定规则将主观题答题区绝对坐标和涂卡分数区的相对坐标信息记录到数据库服务器中。 Under the sideline to answer subjective questions coordinates of each area as a benchmark, to continue using the projection method to obtain relative coordinate information Tuka score area, according to certain rules to answer subjective questions relative coordinate information absolute coordinate area and Tuka scores zone records to the database server. f2.考生答巻信息识别,是识别出考生答巻的填涂信息。 f2. Volume A candidate identification information, is identified in Volume A are marked on the candidate information. 这是在标准答巻图像信息识别完成的基础上进行的,首先将倾斜的考生答巻旋转并校正到与标准答巻图像统一的坐标系上,然后根据位置标准答巻图像中各部分的坐标获取各个识别项的位置信息,以识别出考生答巻的填涂信息。 This is the standard A based Volume-image information identification completion performed on the first inclined candidates A. Volume rotated and corrected to standard A on Volume image unified coordinate system, the coordinates Volume image of each part and then A. The position of the standard acquiring position information of each identification item, information to identify candidates Tiantu Volume of a. 步骤g2的具体步骤如下:f2. 1.考生答巻图像读取与预处理,是将TIFF格式的考生答巻分解成BMP格式的答巻。 Step g2 of the following steps:. F2 1. A candidate image reading Volume pretreatment, the candidate is TIFF format A A Volume Volume decomposed into BMP format. 考生答巻是以TIFF格式的图像包读入的,包含了多张答巻图像。 Candidates A Volume is an image packet read TIFF formats, including multiple A Volume image. 在识别考生信息之前,需先将某个考生的答巻的图像数据从TIFF图像包中分离出来,读入内存并转化成256灰度的BMP格式图像,之后对腦P格式的答巻图像进行去噪、二值化、增强等预处理。 Before identifying candidate information, to be isolated in the first image data of a Volume A TIFF image candidates from out of the package, read into memory and converted to a BMP format image of 256 gradations, Volume A. After the image format is Brain P denoising, binarization, to enhance pretreatment. f2.2.考生答巻图像倾斜校正。 F2.2. Volume A. Candidates image tilt correction. 扫描得到的图像通常会有一些倾斜,本系统的答巻信息识别模块首先识别出考生答巻中的标记位坐标,并以标准答巻图像的标记位坐标为标准,对考生答巻进行旋转和平移,使得考生答巻图像与标准答巻图像位于同一个—坐标豕中。 Scanned images usually have some inclination, the system A first module identification information Volume candidates identified in Volume A flag coordinates and the coordinates of the standard marker bit at Volume A standard image, rotate candidates A and Volume pan, so that the standard image candidates a a Volume Volume in the same image - in hog coordinates. 主要步骤为:首先,使用模板匹配算法获取考生答巻的标记位坐标;其次,根据三个标记位的相对位置判断是否为倒置答巻(若左上、左下、右上有标记位而右下无标记位,则说明答巻是正置的;若左下、右上、右下有标记位而左上无标记位则说明答巻是倒置的。后种情况需要对倒置答巻旋转180度让其正置);然后,根据左上和右上的标记位是水平的先验值来计算倾斜角度,并以此角度进行旋转;最后,将标准答巻标记位坐标与旋转后考生答巻标记位坐标相比较,得到水平和竖直距离差,而后进行相应的平移,使得考生答巻与标准答巻各部分坐标重合。 The main steps: first, using a template matching algorithm acquisition flag Volume A coordinate candidates; secondly, the relative position of the three is determined whether the flag A is inverted Volume (if the upper left, lower left, upper right and lower right bit marked unmarked bits, then the answer is positive opposing Volume; if lower left, upper right, upper left and lower right flag has not then the flag a is inverted Volume latter case the Volume a need inverted upright allowed rotated 180 degrees).; then, according to the flag bit is the upper left and right horizontal priori to calculate the tilt angle, and thus the rotation angle; and finally, after the Volume a standard coordinate with the rotation flag flag coordinate candidates Volume a comparison level to give and a vertical distance difference, then the corresponding translation candidates so that the Volume a and Volume a standard overlap portions of coordinates. f2.3.考号识别,是识别出考生填涂的考号值。 F2.3. test identification number, it is to identify the candidates Tiantu value of candidate number. 考号识别是考生答巻信息识别的第一步,每张答题卡切割的各部分图像是以考号命名区分的。 Candidate number identification is the first step to identify candidates A Volume of information, each answer sheet cutting parts of the image are named after distinguished candidate number. 首先根据标准答巻的考号区域坐标,设定识别范围;向Y轴投影并将投影值与阈值对比识别填涂区域,再根据标准答巻已知各数字的坐标判断填涂数字;之后将识别出的所有数字格式化为考号并返回;最后将考号区域图像切割、索引并保存,以备结果展示和分数核查时使用。 First Volume A standard candidate number region coordinate, the set recognition range; and the projection values ​​and the recognition threshold value comparison in the Y-axis are marked on the projection area, and then according to standard coordinates Volume A digital numbers known Analyzing TianTu; after All digitally formatted identified as candidate number and returns; finally cutting candidate number region image, and stores the index, used to prepare and display the results of verification scores. f2.4.客观题选项识别,是识别出客观题的具体选项。 f2.4. Options recognition objective questions, the objective is to identify the specific question options. 首先,根据标准答巻的客观区域坐标,设定每个题的识别范围;然后,向X轴投影并将投影值与阈值对比从而识别填涂区域;继而,根据标准答巻已知各选项的坐标判断具体填涂选项;之后,将每个题的选项格式化为答案并返回;最后,将整个客观题区域图像切割、索引并保存, 以备结果展示和分数核查时使用。 First, based on objective criteria area coordinate Volume A, the range of each title set identifier; then, comparing the X-axis projection and the projection values ​​with a threshold value to identify Tiantu region; then, according to known criteria A for each option Volume Tiantu options to determine the specific coordinates; then, the formatting options for each question to answer and return; and finally, the entire area image cutting objective questions, indexed and stored, to prepare for verification results show scores and use. f2.5.主观题分数识别,是识别出每个主观题的得分。 f2.5. Subjective question Score identification is to identify the scores for each subjective question. 主观题分数是以涂卡方式记录的,根据标准答巻可知各个主观题答题区的坐标值,再根据涂卡区的相对坐标信息设定识别范围,分别向X轴、Y轴投影并将投影值与阈值对比从而判断填涂位置,根据标准答巻已知的各数字坐标可以识别出所填涂的具体数字;之后将识别的每个数字格式化为分数返回;最后将该主观题及其分数区域的图像切割、索引并保存,以备结果展示和分数核查时使用。 Tuka fraction subjective questions is recorded, according to the standard coordinate value found A Volume answer subjective questions of each region, according to re-setting the relative coordinate recognition range Tuka information area, respectively, in the X-axis, Y-axis projection and the projection comparison with a threshold value to determine Tiantu position, according to the standard known Volume A. each numerical coordinates may identify the particular figures Tiantu; after each number is formatted to identify return points; Finally, the subjective question and fractions the cutting region of the image, and saves the index, used to prepare and display the results of verification scores. f2.6.识别结果返回,是返回一份完整答巻的填涂信息。 f2.6. recognition results are returned, the return of a complete A Volume Tiantu information. 当以上所有信息识别结束后,把所有的信息记录到数据库服务器中,以备结果展示和査询使用。 When all the above information for identifying, recording all the information to the database server in order to prepare and display the results of queries. g. 答巻识别结果校正:是指利用系统的答巻识别校正模块,将识别错误的信息进行校正。 . Volume recognition result correction G A: A refers to the use of system identification Volume correction module, the identification information of the error correction. 为了提高系统的准确性,系统在批量评巻结束后,阅巻教师可以选择任意一个TIFF考生考巻包,察看学生的评分情况,进行误差修正。 In order to improve the accuracy of the system, evaluation system after the end of the batch Volume, Volume reading teacher can choose any of the candidates to test a TIFF Volume package, look at the students' scores case, the error correction. 利用图像切割算法对考生答巻图像进行了分割,分割成为考号区域图像、考生信息区域图像(如姓名、学校和班级等)、客观题区域图像以及各个主观题区域图像。 Cutting algorithm using the image of the candidates A Volume image segmentation, segmentation becomes, objective questions, subjective questions and the various regions of the image area of ​​the image area of ​​the image candidate number, candidate information area of ​​an image (such as name, school and class, etc.). 答巻识别校正模块提供了各个切割图像以及相应的答案识别(针对客观题)或评分(针对主观题)结果,用户可以据此实现校正,以获得准确的数据。 A Volume correction module provides identification of each cut image and the corresponding answers recognition (for objective questions) or scores (for subjective question) As a result, customers can implement a correction to obtain accurate data. gl.修正考号,当识别出的考号出现识别不全、考号重复、考号不存在等异常问题时,问题考号将自动被列出,并根据数据库中的记录将相应的考号图像显示出来,阅巻教师可方便地根据考号图像手工修正考号。 gl. amended candidate number, when the identified test identification number appears incomplete, repeat the test number, the abnormal candidate number is not a problem and the like, problems will automatically be listed in the test number, and the record in the database corresponding image test number displayed, read Volume teachers can easily candidate number manually corrected image based on candidate number. g2.修正客观题分数,客观题也会将其整个题目图像显示出来,以供阅巻教师参考修正。 g2. Fixed score of objective questions, objective questions will be entitled entire image is displayed, read Volume for teachers' reference correction. 当客观题选项异常,如本应单选但考生多选或未选,该题答案会显示异常颜色,鹏,,教,歸7WlOTWMWir实T^W翻nE^巻錢。 When the objective questions options abnormalities, such as radio, but this should be multiple choice or election candidates, the question the answer will show unusual colors, Peng ,, teach, go 7WlOTWMWir real turn nE ^ T ^ W Volume of money. g3.修正主观题分数,当主观题分数出现未识别、分数越界这些异常时。 g3. Fixed score of subjective questions, subjective questions when the score appeared unidentified score out of bounds these anomalies. 系统也会将主观题相应的题目图像显示出来,以供阅巻教师参考修正。 The system will also subjective questions appropriate subject image is displayed, read Volume for teachers' reference correction. h. 自动评分汇总模块:是根据数据库中的考号信息,自动判断考号的有效性,包括考号是否识别完全,考号是否存在,考号是否重复出现等情况;对于主观题,根据当前题目的满分值判断教师所判分数是否合理等。 . H automatic scoring summary module: No test is based on information in the database, automatically determine the number of test validity, including whether the candidate number to identify completely, whether there is candidate number, candidate number whether duplication occurs; for subjective questions, based on the current subject teachers out of value judgment sentence is reasonable score. 之后将各部分得到的分数进行汇总, 从而得出整份答巻的最终分数。 The fractions obtained after the aggregated score to arrive at the final score of the entire Volume A. 本发明的图像处理所采用的关键技术-基于VBA的试巻制作工具是在MS Word基础上做的二次开发,利用了VBA的宏编程技术,能够自动生成标记位,并可根据用户与系统的交互自动地确定答巻的页数、每页的栏数、客观题的数目及主观题的数目与评分区域,使得试巻制作操作简单,所产生的标准答巻图像准确有效。 The key image processing techniques employed in the present invention - again based on VBA Volume Builder MS Word is done on the basis of the second development, the use of the VBA macro programming, can automatically generate flag bits, according to user and system interaction automatically determine the number of pages Volume a, the number of columns per page, the number of objective and subjective question of the number of questions and scoring area, try making Volume production is simple, standard produced by A. Volume image accurately and efficiently. 标记位识别是系统阅巻的关键部分和信息正确识别的基本前提,本系统采用模板匹配技术来实现标记位的识别,并利用标记位实现考生答巻的旋转、平移校正。 Flag bit is the basic premise of the system to identify the key part and read Volume identification information is correct, the system uses the template matching technology to achieve recognition mark bit and effect rotation of the candidates using the marker bit of Volume A translation correction. 对边框、填涂等信息采用投影方法与阈值比较方法,向X轴或Y轴投影,并将此投影值与事先设定的阈值对比,从而判断是否有边框或者是否填涂值。 On the border, and other information are marked on the projection method using a threshold comparison method, projected in the X-axis or Y-axis, and the projection of this comparison with a threshold value preset to determine whether there is a border or whether Tiantu value. 该方法简单、快速、有效,保证了系统的性能。 The method is simple, fast, effective, ensure the performance of the system.

Claims (8)

  1. 1、一种自动化的扫描阅卷系统,其特征在于,由试卷定义模块、基于VBA的答卷制作工具、文档扫描仪、数据库服务器、答卷信息自动识别模块、答卷识别结果校正模块、自动评分汇总模块组成;整个自动阅卷系统是一个星形结构,数据库服务器是系统的中心,其他各个模块及设备都单独以网络的形式与其相连接,用于信息传递和网络共享。 1, an automated scanning system marking, characterized in that the paper definition module, based on the responses VBA Maker, document scanner, database server, automatic answer information identification module, the recognition result correction module respondents, automated scoring summary module ; whole automatic scoring system is a star-shaped configuration, the database server is the center of the system, and each of the other modules in the form of separate devices connected to its network, and for transmitting network sharing information.
  2. 2、 按照权利要求1所述的系统,其特征在于,所述的试巻定义模块安装在用户的计算机上,用于试巻基本信息、结构信息以及答案与评分标准的定义,并把设置结果存储到数据库服务器中,为答巻制作提供信息。 2. The system of claim 1, wherein said sample Volume definition module installed on the user's computer, Volume test for basic information, configuration information, and defines the score of the answer, and the setting result stored in the database server, providing information to answer Volume production.
  3. 3、 按照权利要求1所述的系统,其特征在于,所述的基于VBA的答巻制作工具安装在用户的计算机上,用于生成答巻的电子模板,其输出结果转化为标准答巻图像和考试实际使用的答巻。 3. The system of claim 1, wherein the VBA is based Volume A Builder installed on the user's computer, for generating an electronic template Volume A, the output image into a standard A Volume a test and the actual use of the Volume.
  4. 4、 按照权利要求1所述的系统,其特征在于,所述的文档扫描仪安装在扫描人员的设备间,用来接收大量的经过主观题评阅的考生答巻,对其电子化并自动转存到数据库服务器中。 4. The system of claim 1, wherein said document scanner is mounted between the scanning apparatus personnel, for receiving a large number of candidates through subjective tests reviewers Volume A, and automatically transfer its electronic to keep the database server.
  5. 5、 按照权利要求1所述的系统,其特征在于,所述的答巻信息自动识别模块安装在用户的计算机上,它从数据库服务器中获取考生答巻,之后进行自动识别,最后将识别结果返回给数据库服务器。 5. The system of claim 1, wherein said identification module A. Volume information is automatically installed on the user's computer, it gets Volume candidates from the database server A, after automatic recognition, the final recognition result returned to the database server.
  6. 6、 按照权利要求1所述的系统,其特征在于,所述的答巻信息校正模块,安装在用户的计算机上,从数据库服务器获取自动阅巻的结果信息并显示,供考务人员校正,并将校正结果返回存储到数据库服务器上。 6. The system of claim 1, wherein said correction module A. Volume information, installed on the user's computer, the information is automatically read Volume acquisition result from the database server and displayed for examination personnel correction, and correcting the result stored back to the database server.
  7. 7、 按照权利要求1所述的系统,其特征在于,所述的自动评分汇总模块安装在用户的计算机上,从数据库获取识别校正的结果数据,并根据试巻定义存储的信息, 进行自动的评分与汇总;打印机是系统的可选组件,与自动评分汇总模块所在的计算机相连接,用于打印输出成绩等统计数据。 7. The system of claim 1, wherein said automatic scoring summary module is installed on the user's computer, the corrected recognition result data acquired from the database, and the information stored in the test defined Volume, the automatic summary scores; the printer is an optional component of the system, and a computer module located summary score automatically connected, and the like for printout performance statistics.
  8. 8、 一种采用权利要求1所述的自动化扫描阅巻系统阅巻的方法,其特征在于, 包含以下步骤:a.试巻定义,定义试巻的结构,并进行答案与评分标准的设置, 试巻结构定义,是出题教师利用本系统的试巻定义模块,定义试巻的结构;首先, 需要创建一份试巻的基本信息,包括满分值、考试年级和考试科目;其次,创建题目的数目、各个题目的主客观性质和各个题目的满分值,并把相应的信息存入数据库服务器中;答案与评分标准设置,是指利用本系统的试巻定义模块,在完成试巻结构的定义后,在系统中录入各个客观题的标准答案以及多选题的评分标准,而后试巻定义模块会将这些信息存入数据库服务器中;b. 基于VBA的答巻制作,是指利用本系统提供的基于VBA的试巻制作工具, 根据已定义的试巻结构制作出答巻的电子模板,并把答巻模版存入数 8. An automated method of claim scanning read using Volume Volume reading system according to claim 1, characterized in that it comprises the following steps: a test Volume structure definition, the definition of Volume test, and the score of the answer set. Volume test structure definition, is the topic and teachers to use the system definition module test Volume, Volume defined test structure; first of all, you need to create a test Volume of basic information, including out value, exam grades and test subjects; secondly, to create a title the number of objective and subjective nature of each subject and out of the respective values ​​of the subject, and the corresponding information stored in the database server; score setting answers, refers to the use of this test system definition module Volume, Volume completed the trial structure Once defined, the entry standard answer each multiple choice objective questions and the scoring criteria in the system, then try Volume definition module the information will be stored in the database server;. b a Volume production based on VBA, it refers to the use of the system VBA-based test provides Volume production tools, production structure has been defined according to test Volume Volume a of the electronic template and the template number into Volume A. 库服务器中; 己经制作好的电子模板主要用于以下两个方面: 一是生成标准答巻图像,二是作为考生的答巻;c. 考试,是指考生根据考试安排参与考试的过程;考试之前,考务人员需要将考生的考号收集并存储到数据库服务器中;考试过程中,考生需要书写并填涂答巻中的考生信息区域、将客观题的答题结果填涂在答巻的客观题区域以及将主观题的答题结果书写在答巻的主观题答题区域;d. 主观题评阅,是指阅巻人员对考生答巻中主观题评阅的过程,在评阅完成之后, 阅巻人员将主观题的得分填涂在考生答巻的主观题分数涂卡区中;e. 答巻扫描,是指利用文档扫描仪,将主观题评阅过的考生答巻成批扫描成一个个TIFF格式的图像文件,通常同一试室或班级的考生的答巻会被扫描成一个TIFF, 所得的图像存入数据库服务器内,用于后续的考生答巻 Library servers; electronic template already made good mainly for the following two aspects: First generation A standard Volume image, and second, as the candidates answer Volume; c examination, refer to the candidates according to the process of examination arrangements involved the examination of;. before the exam, Proctors need to collect candidates and candidate number stored in the database server; the examination process, candidates need to write and Tiantu a candidate information in the area Volume, the answer will result in the answer objective questions TianTu Volume of objective questions area and will answer the results of subjective questions written on the subjective question the answer region a Volume of;. d subjective question reviewers, is the process of reading Volume personnel for candidates a Volume in subjective question-reviewed, after the reviewers is completed, read Volume staff the score of subjective questions TianTu subjective Tests a Volume fraction Tuka in the candidate region;. e Volume a scan, using the document scanner means, through the subjective questions reviewers candidates Volume a scanning batches into one TIFF format image files, usually in the same class or test chamber candidates Volume a will be scanned into a TIFF, the resulting image is stored in the database server, the subsequent candidates for Volume a 息识别之用;f. 答巻信息识别,是指利用本系统的答巻信息识别模块,识别出标准答巻和考生答巻中标记位、考号、客观题和主观题等区域的信息,并将识别信息存入到数据库服务器中;还包括:标准答巻信息识别,是定位并记录标准答巻图像的各部分包括标记位、考号、客观题区域、各个主观题答题区域和各个主观题涂卡区域的坐标信息,之后答巻信息识别模块将这些坐标信息存储到数据库服务器中;考生答巻信息识别,是识别出考生答巻的填涂信息;这是在标准答巻图像信息识别完成的基础上进行的,首先将倾斜的考生答巻旋转并校正到与标准答巻图像统一的坐标系上,然后根据位置标准答巻图像中各部分的坐标获取各个识别项的位置信息, 以识别出考生答巻的填涂信息;g. 答巻识别结果校正,是利用本系统的答巻识别结果校正模块,将错 Information for identification purposes; F A Volume information identification, is the use of the present system A. Volume information recognition module recognizes that the standard A. Volume and candidates A Volume in the flag information region of the candidate number, objective items and subjective questions like. and the identification information stored in the database server; further comprising: Volume a standard identification information, and a positioning portions of Volume a recording standard image includes flag, test number, objective questions regions, each region and each answer subjective questions subjective Tuka title area coordinate information, then the information recognition module A. Volume coordinates information into the database server; Volume a candidate identification information is identification information of a candidate Tiantu Volume a; this is the standard-image information identification Volume a performed Upon completion of the first inclined candidates A. Volume rotated and corrected to the standard a Volume image unified coordinate system, then a coordinate Volume image of each part acquires position information of each identification item according to the location criteria to a candidate identified Tiantu Volume of information;. g Volume a recognition result correction, using this system module A. Volume recognition result correction, the error 的识别信息进行校正;对考生答巻图像分割成为考号区域图像、考生信息区域图像、客观题区域图像以及各个主观题区域图像;根据数据库服务器中的考号信息,自动判断已识别考号的有效性:包括考号是否识别完全、考号是否存在、考号是否重复出现等情况;对于主观题,根据当前题目的满分值判断教师所判分数是否合理;针对各种不合理的情况,给出友好明确的提示;答巻识别结果校正模块提供了各个切割图像以及相应的客观题答案识别或主观题评分结果,阅巻人员据此实现校正,校正后的信息将存储到数据库服务器中;h.自动评分汇总,是利用本系统的自动评分汇总模块,根据自动识别和校正后的客观题答案与主观题评分识别结果,依据评分规则,自动评分汇总每份答巻的分数。 The identification information is corrected; examinee A Volume segmentation become candidate number region image candidate information region of an image, an objective problem area image and each of the subjective question region image; The candidate number information in the database server automatically determines the identified test number effectiveness: including whether the candidate number to identify completely, whether there is candidate number, candidate number, etc. if repeated; for subjective questions, according to the judge to determine the value of teachers out of the current title score is reasonable; unreasonable for a variety of circumstances, to a friendly and clear tips; a Volume recognition result correction module provides various cutting image and the corresponding objective questions, subjective questions or answers recognition score results, read Volume personnel pursuant to realize correction information after the correction will be stored in the database server; h automatic scoring summary, this system is the use of automatic scoring summary module, based on objective answer questions after the automatic identification and correction of problems with subjective scoring recognition result, according to the scoring rules, automatic scoring Volume a summary of each score.
CN 200810105224 2008-04-28 2008-04-28 Automatic scanning and reading system and reading method CN101334814A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 200810105224 CN101334814A (en) 2008-04-28 2008-04-28 Automatic scanning and reading system and reading method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 200810105224 CN101334814A (en) 2008-04-28 2008-04-28 Automatic scanning and reading system and reading method

Publications (1)

Publication Number Publication Date
CN101334814A true true CN101334814A (en) 2008-12-31

Family

ID=40197411

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 200810105224 CN101334814A (en) 2008-04-28 2008-04-28 Automatic scanning and reading system and reading method

Country Status (1)

Country Link
CN (1) CN101334814A (en)

Cited By (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101901338A (en) * 2010-07-09 2010-12-01 北京商纳科技有限公司 Method and system for calculating scores of test paper
CN102339352A (en) * 2010-07-20 2012-02-01 上海海鞠电子科技有限公司 Electronic paper marking method
CN102663379A (en) * 2012-03-22 2012-09-12 留越 Paper marking method based on image identification and system thereof
CN102682307A (en) * 2012-05-03 2012-09-19 苏州多捷电子科技有限公司 Modifiable answer sheet system and implementation method thereof based on image processing
CN102855495A (en) * 2012-08-22 2013-01-02 苏州多捷电子科技有限公司 Method for implementing electronic edition standard answer, and application system thereof
CN102890775A (en) * 2011-07-21 2013-01-23 北京五岳鑫信息技术股份有限公司 Test score automatic recognition processing method based on optical character recognition (OCR)
CN102968627A (en) * 2011-08-31 2013-03-13 上海夏尔软件有限公司 Precise image segmentation method based on area keyword position recognition
CN103279907A (en) * 2013-05-13 2013-09-04 南昊(北京)科技有限公司 Online-scanning paper inspection system
CN103310082A (en) * 2012-03-07 2013-09-18 爱意福瑞(北京)科技有限公司 Paper inspection method and device
CN103325078A (en) * 2013-05-15 2013-09-25 北京环宇锦盛科技有限公司 On-site examination mobility and controllability marking method and system based on tablet computers
CN103559515A (en) * 2013-10-24 2014-02-05 江苏玖宇实业有限公司 Intelligent scoring pen
CN103745104A (en) * 2013-12-31 2014-04-23 广东工业大学 Examination paper marking method based on augmented reality technology
CN103778816A (en) * 2014-01-27 2014-05-07 上海五和文化传播有限公司 Test paper answer extracting system
CN104143082A (en) * 2014-07-08 2014-11-12 北京彩云动力教育科技有限公司 Test paper information positioning method and system based on image fuzz recognition
CN104239866A (en) * 2014-09-16 2014-12-24 科大讯飞股份有限公司 Answer sheet information collection method and device
CN104658349A (en) * 2015-02-06 2015-05-27 濮阳职业技术学院 Art teaching system
CN105069412A (en) * 2015-07-27 2015-11-18 中国地质大学(武汉) Digital scoring method
CN105096677A (en) * 2015-08-19 2015-11-25 北京京东方多媒体科技有限公司 Teaching system and work method thereof
CN105335957A (en) * 2014-08-07 2016-02-17 江苏奥博洋信息技术有限公司 Information fragmentization picture cutting method
CN105426879A (en) * 2015-12-23 2016-03-23 深圳习习网络科技有限公司 Method and system for identifying answer sheet
CN105718438A (en) * 2016-03-04 2016-06-29 北京学信速达科技有限公司 Typesetting style generation method for general answer sheet
CN105740854A (en) * 2016-05-16 2016-07-06 苏州金建达智能科技有限公司 Intelligent reading and amending system for teaching
CN105894880A (en) * 2016-05-12 2016-08-24 湖南汇博电子技术有限公司 Method for answering questions through network and server
CN103955889B (en) * 2013-12-31 2017-02-08 广东工业大学 Augmented reality mapping technology reviewers based on class work
CN107451581A (en) * 2017-07-11 2017-12-08 济宁九点连线信息技术有限公司 Method for automatically approving answer sheets in batch

Cited By (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101901338A (en) * 2010-07-09 2010-12-01 北京商纳科技有限公司 Method and system for calculating scores of test paper
CN102339352A (en) * 2010-07-20 2012-02-01 上海海鞠电子科技有限公司 Electronic paper marking method
CN102890775A (en) * 2011-07-21 2013-01-23 北京五岳鑫信息技术股份有限公司 Test score automatic recognition processing method based on optical character recognition (OCR)
CN102968627A (en) * 2011-08-31 2013-03-13 上海夏尔软件有限公司 Precise image segmentation method based on area keyword position recognition
CN103310082A (en) * 2012-03-07 2013-09-18 爱意福瑞(北京)科技有限公司 Paper inspection method and device
CN102663379A (en) * 2012-03-22 2012-09-12 留越 Paper marking method based on image identification and system thereof
CN102682307A (en) * 2012-05-03 2012-09-19 苏州多捷电子科技有限公司 Modifiable answer sheet system and implementation method thereof based on image processing
CN102855495A (en) * 2012-08-22 2013-01-02 苏州多捷电子科技有限公司 Method for implementing electronic edition standard answer, and application system thereof
CN103279907A (en) * 2013-05-13 2013-09-04 南昊(北京)科技有限公司 Online-scanning paper inspection system
CN103325078A (en) * 2013-05-15 2013-09-25 北京环宇锦盛科技有限公司 On-site examination mobility and controllability marking method and system based on tablet computers
CN103559515A (en) * 2013-10-24 2014-02-05 江苏玖宇实业有限公司 Intelligent scoring pen
CN103745104A (en) * 2013-12-31 2014-04-23 广东工业大学 Examination paper marking method based on augmented reality technology
CN103745104B (en) * 2013-12-31 2017-03-29 广东工业大学 Augmented reality technology based marking
CN103955889B (en) * 2013-12-31 2017-02-08 广东工业大学 Augmented reality mapping technology reviewers based on class work
CN103778816A (en) * 2014-01-27 2014-05-07 上海五和文化传播有限公司 Test paper answer extracting system
CN104143082A (en) * 2014-07-08 2014-11-12 北京彩云动力教育科技有限公司 Test paper information positioning method and system based on image fuzz recognition
CN104143082B (en) * 2014-07-08 2018-06-05 北京彩云动力教育科技有限公司 Kinds of paper information based positioning method and system for image blur Recognition
CN105335957A (en) * 2014-08-07 2016-02-17 江苏奥博洋信息技术有限公司 Information fragmentization picture cutting method
CN104239866B (en) * 2014-09-16 2018-01-12 科大讯飞股份有限公司 Answer sheet information collection method and apparatus
CN104239866A (en) * 2014-09-16 2014-12-24 科大讯飞股份有限公司 Answer sheet information collection method and device
CN104658349A (en) * 2015-02-06 2015-05-27 濮阳职业技术学院 Art teaching system
CN105069412A (en) * 2015-07-27 2015-11-18 中国地质大学(武汉) Digital scoring method
CN105096677A (en) * 2015-08-19 2015-11-25 北京京东方多媒体科技有限公司 Teaching system and work method thereof
CN105426879A (en) * 2015-12-23 2016-03-23 深圳习习网络科技有限公司 Method and system for identifying answer sheet
CN105718438A (en) * 2016-03-04 2016-06-29 北京学信速达科技有限公司 Typesetting style generation method for general answer sheet
CN105894880A (en) * 2016-05-12 2016-08-24 湖南汇博电子技术有限公司 Method for answering questions through network and server
CN105740854A (en) * 2016-05-16 2016-07-06 苏州金建达智能科技有限公司 Intelligent reading and amending system for teaching
CN107451581A (en) * 2017-07-11 2017-12-08 济宁九点连线信息技术有限公司 Method for automatically approving answer sheets in batch
CN107451581B (en) * 2017-07-11 2018-08-21 山东九点连线信息技术有限公司 Kind of automatic batch marking the answer sheet approach

Similar Documents

Publication Publication Date Title
Laurent The cultural diversity of western conceptions of management
Hayes et al. Cultural values, plagiarism, and fairness: When plagiarism gets in the way of learning
US20050268217A1 (en) System for describing the overlaying of electronic data onto an electronic image
US5672060A (en) Apparatus and method for scoring nonobjective assessment materials through the application and use of captured images
Phuntsog Culturally responsive teaching: What do selected United States elementary school teachers think?
US6970267B1 (en) Gray scale optical mark reader
US6042384A (en) Computerized systems for optically scanning and electronically scoring and reporting test results
Li Tailoring translation programs to social needs: a survey of professional translators
Love et al. Designing online information systems for portfolio-based assessment: Design criteria and heuristics
US20070178432A1 (en) Test management and assessment system and method
US20100075290A1 (en) Automatic Educational Assessment Service
Goldstein et al. Can Students Benefit from Process Writing?.
Cousins et al. The nature and impact of policy-induced partnerships between research and practice communities
Campbell Application of ICT and rubrics to the assessment process where professional judgement is involved: the features of an e‐marking tool
Walker Typography & Language in Everyday Life: Prescriptions and Practices
Crossley et al. Textbook provision and the quality of the school curriculum in developing countries: Issues and policy options
Soper Test of Economic Literacy: Discussion guide and rationale.
Nelson et al. The school-wide cultural competence observation checklist for school counselors: An exploratory factor analysis
CN1598845A (en) Network exanmination paper go-over method for subjective question
Rafoth et al. The impact of content and mechanics on judgments of writing quality
Supovitz et al. Assessing agreement: An examination of the interrater reliability of portfolio assessment in Rochester, New York
Lord et al. Foreign language teacher preparation and asynchronous CMC: Promoting reflective teaching
US20060045344A1 (en) Handprint recognition test deck
Lecouteur et al. Repertoires of teaching and learning: A comparison of university teachers and students using Q methodology
Sless What is information design

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination
C02 Deemed withdrawal of patent application after publication (patent law 2001)