当前位置：文档库 › MAP-regularized robust reconstruction for underwater imaging detection

MAP-regularized robust reconstruction for underwater imaging detection

Optik124 (2013) 4514–4518

Contents lists available at ScienceDirect

Optik

j o u r n a l h o m e p a g e:w w w.e l s e v i e r.d e/i j l e

MAP-regularized robust reconstruction for underwater imaging detection Yuzhang Chen a,?,Kecheng Yang b

a Faculty of Physics and Electronic Technology,Hubei University,Wuhan430062,China

b College of Optoelectroni

c Science an

d Engineering,Huazhong University of Scienc

e and Technology,Wuhan430074,China

a r t i c l e i n f o

Article history:

Received16August2012

Accepted14January2013

Keywords:

Information capacity

MAP-regularized

Point spread function

Robust super-resolution reconstruction Underwater imaging detection a b s t r a c t

In order to enhance the visual quality of underwater images,applications such as enhancement and restoration can be applied,but the resolution is still limited.Super-resolution reconstruction is a widely used technique for improving resolution beyond the limit of imaging system.With knowledge of the point spread function and techniques of regularization,the performance of reconstruction can be fur-ther enhanced.The presented effort proposed a robust image super-resolution reconstruction method under maximum a posteriori framework with regularization by the point spread function for underwater imaging detection.Objective image quality metrics are used to quantify the effectiveness of the recon-struction.Experimental results showed that the proposed method can effectively improve the resolution and quality of underwater imaging detection.

1.Introduction

Underwater imaging which is widely used in ocean exploration has been developed since1963when a transmission window of 470–580nm in underwater environment was found by Duntley and Gilbert[1].However,the resolution of underwater imaging is limited by both the optical properties such as absorption,scattering effects within formation medium and the system design includ-ing light sources and image sensors.Over past years,efforts such as image restoration methods has been studied and proposed by researchers such as Weilin Hou[2],Yifan Yu[3],Fan Fan[4].Image restoration methods can effectively eliminate the blur caused by absorption and scattering;however,the effectiveness for improv-ing resolution is still constraint[5].

Image super-resolution reconstruction which has been increas-ingly popular ever since Tsai and Huang[6]proposed the concept in1984offers the possibility of improving image resolution beyond the hardware limitations.It refers to achieving high-resolution(HR)images from one or multiple low-resolution(LR) images[7]using the complementary information between image sequences.Image registration transfers low-resolution images into non-uniform samples[8],which then will be processed by image fusion using interpolation or other methods to high-resolution images.

The robust super-resolution reconstruction(RS)method is a rel-atively new method proposed in2001[9].It is achieved by the combination of a robust median estimator and an iterative process

?Corresponding author.

E-mail address:syz1984727@https://www.wendangku.net/doc/1010131755.html,(Y.Chen).in the purpose of minimizes the error caused by model inaccuracies and moving objects,etc.The blurring factor in the calculation of this method can be estimated by the optical properties of the imaging system.The PSF of an imaging system gives the system response including the optical system[10],as a result,it is an intuitive and ideal parameter to describe the retrieval of optical properties and help to break the resolution limitation by hardware.For the error minimization process,a statistical regularization algorithm based on maximum a posteriori(MAP)[11]provides a convenient frame-work under which image reconstruction can be performed to a best solution.

As a result,considering the technique of the robust super-resolution reconstruction algorithm,we can derive the PSF from empirical point spread models?rst[12]and then enhance the resolution by the MAP-regularized RS method.Thus,the appli-cations of the proposed method for underwater imaging are compared and validated by image evaluation in this presented effort.

2.Theory

2.1.Obtaining of the point spread function

The purpose of point spread function(PSF)is to predict the image intensity at each pixel as a function of illumination, re?ectance properties of objects,medium,and sensor character-istics.The degradation by imaging system including the diffraction limit of optics and nonlinear distortion of sensors can help to?nd a way for breaking through the hardware limitations.As a result, obtaining the PSF of degradation by imaging system is an important work for super-resolution reconstruction.

Y.Chen,K.Yang /Optik 124 (2013) 4514–4518

4515

When only one main lens exists in the optical system,the mod-ulation transfer function (MTF)of diffraction limit can be calculated as [13]:

MTF diff

arccos

f f co

(1)

f co =

D p ×f l

(2)

where f is the spatial frequency,f co denotes the optical cut off fre-quency at the image plane,f l denotes the focal length,D p is the diameter of the lens,and is the wavelength of operation.The mod-ulation transfer function of the CCD sensor depends on the size of pixels,which can be de?ned as:

MTF ccd =

sin( ×d pixel ×f )

×d pixel ×f

(3)

where d pixel is the size of pixels,f denotes the spatial frequency.

Therefore,the MTF of degradation can be obtained by mul-tiplying the components caused by diffraction of the optics and nonlinear distortion of sensors:

MTF img =MTF diff ×MTF ccd =

arccos

f f co

sin( ×d pixel ×f )

×d pixel

(4)

the PSF can be obtained from MTF using Hankel transform [14]:

h (?,L )=2

J 0(2 ??)H (?,L )?d?

(5)

where h (?,L )denotes the PSF of optical system,and H (?,R )for the

MTF,J 0is the Bessel function of order zero,?is the displacement from the unperturbed beam axis,?is the spatial frequency in cycles per radian,and L is the imaging distance.The PSF which gives the system response is an intuitive way for describing the image forma-tion and the limitation of imaging system,so it can help to enhance the performance of super-resolution reconstruction.As a result,it is used for the MAP-regularized RS method.

2.2.MAP-regularized RS method

Registration is of vital importance before super-resolution reconstruction,it is a mapping operation between two images spatially for projecting onto the high-resolution grid,including algorithms in frequency domain [15]and spatial domain [16].The Keren registration [17]is applied for underwater imaging for its accuracy and robustness.The registration operation between two images g (x ,y )and f (x ,y )is based on a rigid body transformation model:

g (x,y )=f (x cos ??y sin ?+a,y cos ?+x sin ?+b )

(6)

where a is horizontal shift,b is vertical shift,and ?is the rotation angle.

The main idea of the RS method has some resemblance to that of the iterative back-projection (IBP)method.It can be expressed by an iterative equation:

f n +1=f n + ?L (f )

(7)

?L (f )=n ·median {HB k }n k =1=nH ·median {B k }n

k =1

(8)

where denotes the scale factor of the gradient step size,k is the number of low-resolution image frames.f n +1and f n denote the results from (n +1)th iteration and n th iteration,respectively,B k represents the back-projected difference image, L (f )is equal

to a scaled pixel-wise median with the purpose of introducing robustness into super-resolution reconstruction.A median can approximate the mean quite accurately for a symmetric distribu-tion,given a suf?cient set of samples;so it is much more robust than the mean.H is the blurring operator,the choice of H affects the characteristics of the solution when there are possible solutions;thus,it can be utilized as an additional constraint which represents the desired property of the solution [7].As a result,the arbitrarily chosen value of H such as point spread function can determine the appropriate iteration time and extend the performance of the RS method.

The maximum a posteriori (MAP)framework with regulariza-tion by the point spread function can be applied for obtaining optimal solution for the RS method.Considering the most com-mon model used for image formation in which the relation between observed image f and uncorrupted image f can be described as:

f =f ?h +n

(9)

where h is the point spread function of the system,*denotes the convolution operation,n denotes the noise of the system.The main idea of the MAP algorithm is to achieve a best estimation of f by maximizing the posteriori probability:

f map =ar

g max f [P r (f |f )]

(10)

As for multiframes {g i }k i =1

which denote k frames of low resolu-tion,according to the Bayesian theorem,Eq.(10)becomes:

f map =ar

g max f [P r (f |g 1,g 2,...,g i ,...,g k )]

=arg max f [log P r ({g i }k i =1

|f )+log P r (f )](11)

Assigning zero to the derivative of f ,we can have:

?log P r ({g i }k i =1

|f )?f

?log P r (f )

(12)

so the solution of which is the best estimation of f .

The choice for P r (f )depends on the hypothesis for different situ-ations such as Poisson or Markov random distribution.The Markov random distribution is more popular in the application of image reconstruction for its advantage in comprehensive describing the image features,and the Poisson random distribution is included in Markov random distribution by setting its parameters.The Markov distribution is chosen in our method for denoising considerations which has the form:

P r (f )=exp ?

||Qf ||2

(13)

where Q is the linear high-pass operator used to punish non-smoothing estimation and the variance, is the variance control factor.As a result,the Markov random distribution model provides an effective way to describe the image characteristics.By selecting appropriate parameters,it can provide speci?c smoothness con-straint,and effectively improve the effect of image super-resolution reconstruction.From Eqs.(12)and (13),we can derive the MAP-regularized solution:

f map =ar

g min[||f ?hf ||2??||Qf ||2]

(14)

where ?is control factor of regularization,Q denotes the two

dimensional Laplasse operator.The h which denotes PSF can intro-duce the advantage of priori knowledge into the regularization.Above all,from the combination of Eqs.(7)and (14),we can get an optimal solution of the uncorrupted image.

2.3.Image evaluation

Image quality metrics are needed for evaluating the perfor-mance of the point spread model-based RS methods.As no ideal

4516Y.Chen,K.Yang /Optik 124 (2013) 4514–

4518

Fig.1.The framework of underwater range-gated imaging

system.

Fig.2.Sample images of the video frames (size 486×475pixels)at the imaging distance of:(a)35m,(b)40m.

or reference image exists for underwater imaging systems,blind,objective image quality metrics such as gray mean grads (GMG),laplacian sum (LS)and information capacity (IC)are chosen.The GMG and LS proposed by Sheikh and Bovik [18]can effectively re?ect the edge pro?le of an image:

GMG =

(M ?1)(N ?1)

M ?1 i =1N ?1 j =1

[f (x,y +1)?f (x,y )]2+[f (x +1,y )?f (x,y )]

(15)

LS =

M ?1i =1 N ?1

j =1

|8?f (x,y )?f (x,y ?1)?f (x ?1,y )?f (x +1,y )?f (x,y +1)?f (x ?1,y ?1)?f (x ?1,y +1)?f (x +1,y ?1)?f (x +1,y +1)|

(M ?2)(N ?2)

(16)

where f (x ,y )denotes the point at coordinate (x ,y )on image plane,(M ,N )is the size of image.The IC [19]is de?ned as:

C (d,?)=log 2

w log[p (i,j,d,?)]log[max(p (i,j,d,?))]

(17)

where p (i ,j ,d ,?)is the relativity between pixels which has gray levels of i and j ,distance of d ,and direction of ?.It can be seen from above mathematical expressions that the GMG,LS and Information capacity of better images is higher than that of degraded ones.

3.Experimental results and discussion

The experimental data were obtained by an underwater range-gated imaging system which consists of a Q-switch,frequency doubled Nd:YAG laser operated at 532-nm as light source,and an intensi?ed-CCD (ICCD)as image sensor.Fig.1shows the schematic diagram of imaging system with physical properties listed in Table 1.Aligning optic is used to vary the illumination ?eld of view and a programmable timing generator is used as external trigger controller for the purpose of gating.The experiment was conducted in a ship pond,the target was located at different distances from

Table 1

Main physical properties of underwater range-gated imaging system.

Parameters

Value

Relative aperture (D /f l )0.25Imaging wavelength (?)532nm Peak power of laser (P 0)107W Attenuation coef?cient (k )

0.25m ?

Fig.3.MTF img as a function of spatial frequency.

laser and ICCD.The angle of ?eld of view (FOV)is about 4?,which ?ts the range limitation of Wells’theory (0?

When the imaging distance of target is at 35m and 40m,10frames are extracted from the test video sequences collected by ICCD for RS reconstruction.Two sample images are shown in Fig.2for 35m and 40m,respectively,the original image size of video frame is 486×475pixels.By using the parameters of experimental facilities,the curve of MTF img as a function of spatial frequency is shown in Fig.3.

The following algorithms are compared and validated:(1)tradi-tional RS method,(2)traditional MAP method,(3)MAP-regularized RS method.Fig.4shows the compared reconstruction results.The GMG,LS and IC values of the reconstructed results appear in Table 2.

From the pixel size of the original images and reconstructed images in Fig.4,we can see that the improvement of resolution by super-resolution reconstruction is obvious.There exists some blurring effect in the reconstructed results by MAP method,which is eliminated by MAP-regularized RS method due to the Markov random distribution based regularization.From Table 2,the MAP-regularized RS method performs better than either RS method or MAP method.It can be seen that a clear increase in GMG and LS value when PSF and regularization is used.

As a result,it can be concluded that the merge of point spread function and techniques of regularization can enhance the perfor-mance and effect of super-resolution reconstruction.

In order to further demonstrate the effect of proposed method,Canny edge detection is used,the results of which are shown in Fig.5.It also can be seen that the MAP-regularized RS method can substantially enhance the resolution and quality of underwater imaging.

Y.Chen,K.Yang /Optik 124 (2013) 4514–4518

4517

Fig.4.Reconstruction results (size 1458×1425pixels)of images at distance of 35m and 40m,respectively,by:(a1–a2)RS method,(b1–b2)MAP method,(c1–c2)MAP-regularized RS

method.

Fig.5.The edge detection results of images:(a)original 35m,(b)original 40m,(c)MAP-RS-35m,(d)MAP-RS-40m.

T a b l e 2C o m p a r i s o n b e t w e e n t h e r e c o n s t r u c t e d r e s u l t s .

R S -35m

R S -40m

M A P -35m

M A P -40m

M A P -R S -35m

M A P -R S -40m G M G 2,677,9642,699,4362,866,3402,735,24912,567,70312,349,952L S 13,828,58412,973,20312,247,40213,953,943118,626,897112,836,627I C

0.0827

0.0405

0.2607

0.2530

0.5558

0.4379

4518Y.Chen,K.Yang/Optik124 (2013) 4514–4518

As the MAP-regularized RS method do not rely on any speci?c type of imaging system or environment,it can also be bene?cial to various applications.The enhancement of resolution and quality of underwater imaging can be bene?cial to underwater detection which will be the following work of this paper.

4.Conclusions

Point spread function along with regularization techniques are combined to a robust super-resolution reconstruction method for underwater imaging.The results show that the PSF of imaging optics can help to improve resolution beyond the limit of hardware, and the regularization techniques can help to enhance the per-formance of reconstruction.The proposed MAP-regularized robust super-resolution reconstruction method can effectively enhance the resolution and quality of underwater imaging,which is ben-e?cial to underwater detection and other various applications. References

[1]S.Q.Duntley,Light in the sea,J.Opt.Soc.Am.53(1963)214–233.

[2]W.Hou,D.J.Gray,A.D.Weidemann,Automated underwater image restoration

and retrieval of related optical properties,in:International Geoscience and Remote Sensing Symposium(IGARSS),2007,pp.1889–1892.

[3]Y.Yu,F.Liu,System of remote-operated-vehicle-based underwater blurred

image restoration,Opt.Eng.46(2007).

[4]F.Fan,K.Yang,B.Fu,Application of blind deconvolution approach with image

quality metric in underwater image restoration International Conference on Image Analysis and Signal Processing,2010,pp.236–239.

[5]W.Hou,Z.Lee,A.D.Weidemann,Why does the Secchi disk disappear?An

imaging perspective,Opt.Express15(2007)2791–2802.

[6]R.Y.Tsai,T.S.Huang,Multipleframe image restoration and registration,Adv.

Comput.Vis.Image Process.(1984)317–339.

[7]P.Sung Cheol,P.Min Kyu,K.Moon Gi,Super-resolution image reconstruction:

a technical overview,IEEE Signal Process.Mag.20(2003)21–36.

[8]W.Zhao,H.Sawhney,M.Hansen,Super-fusion:a super-resolution method

based on fusion Proceedings16th International Conference on Pattern Recog-nition,2002,pp.269–272.

[9]A.Zomet,A.Rav-Acha,S.Peleg,Robust super-resolution Proceedings of the IEEE

Computer Society Conference on Computer Vision and Pattern Recognition, CVPR,2001,pp.645–650.

[10]H.Weilin,A simple underwater imaging model,Opt.Lett.34(2009)2688–

2690.

[11]S.Huanfeng,L.Pingxiang,Z.Liangpei,A MAP algorithm to super-resolution

image reconstruction,in:Proceedings of Third International Conference on Image and Graphics,2004,pp.544–547.

[12]J.W.McLean,K.J.Voss,Point spread function in ocean water:comparison

between theory and experiment,Appl.Opt.30(1991)2027–2030.

[13]D.Bonnier,S.Chatigny,Y.Lutz,Modelling of active TV system for surveil-

lance operations,Proceedings of SPIE–The International Society for Optical Engineering3698(1999)217–228.

[14]H.Weilin,D.J.Gray,A.D.Weidemann,Automated underwater image restora-

tion and retrieval of related optical properties,in:IEEE International Geoscience and Remote Sensing Symposium,2007,pp.1889–1892.

[15]P.Vandewalle,S.Susstrunk,M.Vetterli,A frequency domain approach to regis-

tration of aliased images with application to super-resolution,EURASIP J.Appl.

Signal Process.2006(2006)14.

[16]J.R.Bergen,P.Anandan,K.J.Hanna,Hierarchical model-based motion estima-

tion,in:Second European Conference on Computer Vision Proceedings,1992, pp.237–252.

[17]D.Keren,S.Peleg,R.Brada,Image sequence enhancement using sub-pixel dis-

placements,in:The Computer Society Conference on Computer Vision and Pattern Recognition,1988,pp.742–746.

[18]H.R.Sheikh,A.C.Bovik,Image information and visual quality,IEEE Trans.Image

Process.15(2006)430–444.

[19]H.Hongwei,Z.Xiaohui,G.Weilong,Performance evaluation of underwater

range-gated viewing based on image quality metric,in:9th International Con-ference on Electronic Measurement Instruments,2009,pp.440–441.

文件和资料标准化管理制度

文件和资料标准化管理制度 1 目的通过对公司文件资料的有效控制，确保工作现场使用唯一有效的文件资料，并形成统一规范的编写格式及处理程序。 2 适用范围本制度适用于公司所有公务文书、文件和资料的管理。 3 术语和定义 3.1 公文：指公司在处理各种公务时使用的应用文书，包括：决定、决议、通知、通报、报告、请示、批复、函（电报）、会议纪要。 3.2 制度（程序）：指要求公司成员共同遵守的，按一定程序办事的规程性文件。 3.3 办法：指公司针对某项工作依照其所需标准制定的考核奖惩性的文件。 3.4 标准：指公司对某项工作应达到的要求进行规范和约定的文件。 3.5 规章制度：公司各种制度、规定、办法的泛称。 3.6 记录：指公司对某项活动的各工作环节、结果进行记录，可供事后追溯该项活动完成质量的证据性文件，主要为表格形式。 4 职责权限 4.1 经理负责公司制度、办法、标准、计划、记录和以公司名义下发的公文的批准。 4.2 管理者代表负责制度、办法、标准、计划、记录的审核。 4.3 各单位负责本单位相关文件资料的编写、审核、打印、校对工作。 4.4 总经办负责以公司名义下发执行的各类文件资料的编号、发放、收回、作废销毁和归档，负责对各单位文件资料管理情况进行监督检查。 4.5 各单位负责以本单位名义下发执行的各类文件资料的编号、发放、收回、作废销毁和归档。 5 工作程序 5.1 文件资料的编写格式 5.1.1 文件资料的用纸标准：公司各类文件资料的正式编印一般用A4(210mm×297mm)型纸张。图纸表格等不宜减小时，该页应按以上纸型尺寸折叠装订。张贴的公文用纸大小，根据实际需要确定。 5.1.2 文件和资料章、条、款的编排规则： 5.1.2.1 根据文件内容的编排划分，“章”就是一个章节，包括条和款，“条”是章的一个部分，“款”是章或条的一个层次。章、条均用阿拉伯数字编号。 5.1.2.2 “章”应左起空两格书写，如第1章用“1”标注，右侧空一格写该章的标题，标题一行书写不够时，可另起一行，这一行的第一个字与标题的第一个字平排书写。章与章之间应空一行编排。 5.1.2.3 “条”在“章”的编号右下加一个小圆点，再写该条在该章所处的顺序，如第4章第1条，用“4.1”表示，后空一格书写该条的标题。如果一章的各条根据需要再分为若干下一层次的小条，其编号表示方法同上，如第4章第2条，再分3小条，则分另表示为“4.2.1”、“4.2.2”、“4.2.3”，如果“4.2.1”再分为2个小条，则分别表

文件资料标准化管理制度

【最新资料Word版可自由编辑！！】文件资料标准化管理制度 1 目的

为进一步加强文件管理，规范文件流程，提高工作效率和办文效率，充分发挥文件上传下达的作用，确保公司政令畅通，特制定本制度。 2 适用范围 2.1 本制度适用于公司文件资料的管理. 2.2 本制度指的文件资料包括公文、管理制度、计划、外来文件和资料等。 3职责权限 3.1行政办负责以公司名义上报、下发执行的各类文件资料的编号、发放、收回、归档和销毁。 3.2 其它各部门负责以部门名义上报、下发的各类文件资料的管理，行政办负责文件资料审查、打印、校对工作。 3.4各部门的对外文件，经行政部文字格式的审核，上报主管副总经理审批。 4 术语定义 4.1 公文——指总公司在处理各种公务时使用的应用型文书，包括：决定、决议、通知、通报、报告、请示、批复、函、会议纪要。 4.2 管理制度——指要求公司成员共同遵守，按一定程序工作、达到一定标准，并进行考核的文件。 4.3 计划——指总公司为完成一定时期内的工作任务，而事先做出安排的文件。 5 工作内容及程序 5.1 文件资料的编写格式 5.1.1 文件资料的用纸标准型纸张。图样表格等不宜减小时，该页必须按以上尺寸折叠装订。一般用型A 4 5.1.2 文件和资料章、条、款的编排规则 5.1.2.1 根据文件内容的编排划分，章就是一个章节，包括条和款，条是章的一个部分，款就是章或条的一个层次。 5.1.2.2 章用阿拉伯数字表示，第一章就用“1”表示，以下按顺序依次类推。“章”应左起空两格书写，右侧空一格写该章的标题，该标题一行书写不够，可另起一行，但这一行的第一个字与该标题的第一个字平排书写。章与章之间应空一行编排。

公司文件和资料标准化管理制度

公司文件和资料标准化管理制度 1．目的通过对公司文件资料的有效控制，确保工作现场使用唯一有效的文件资料，并形成统一规范的编写格式及处理程序。 2．适用范围本制度适用于公司所有公务文书、文件和资料的管理。 3．术语和定义 3．1 公文：指公司在处理各种公务时使用的应用文书，包括：决定、决议、通知、通报、报告、请示、批复、函（电报）、会议纪要。 3．2 制度（程序）：指要求公司成员共同遵守的，按一定程序办事的规程性文件。 3．3 办法：指公司针对某项工作依照其所需标准制定的考核奖惩性的文件。 3．4 标准：指公司对某项工作应达到的要求进行规范和约定的文件。 3．5 规章制度：公司各种制度、规定、办法的泛称。 3．6 记录：指公司对某项活动的各工作环节、结果进行记录，可供事后追溯该项活动完成质量的证据性文件，主要为表格形式。 4．职责权限 4．1 经理负责公司制度、办法、标准、计划、记录和以公司名义下发的公文的批准。 4．2 管理者代表负责制度、办法、标准、计划、记录的审核。 4．3 各单位负责本单位相关文件资料的编写、审核、打印、校对工作。 4．4 总经办负责以公司名义下发执行的各类文件资料的编号、发放、收回、作废销毁和归档，负责对各单位文件资料管理情况进行监督检查。 4．5 各单位负责以本单位名义下发执行的各类文件资料的编号、发放、收回、作废销毁和归档。 5．工作程序 5．1 文件资料的编写格式 5．1．1 文件资料的用纸标准：公司各类文件资料的正式编印一般用a4（210mm×297mm）型纸张。图纸表格等不宜减小时，该页应按以上纸型尺寸折叠装订。张贴的公文用纸大小，根据实际需要确定。 5．1．2 文件和资料章、条、款的编排规则： 5．1．2．1 根据文件内容的编排划分，“章”就是一个章节，包括条和款，“条”是章的一个部分，“款”是章或条的一个层次。章、条均用阿拉伯数字编号。 5．1．2．2 “章”应左起空两格书写，如第1章用“1”标注，右侧空一格写该章的标题，标题一行书写不够时，可另起一行，这一行的第一个字与标题的第一个字平排书写。章与章之间应空一行编排。 5．1．2．3 “条”在“章”的编号右下加一个小圆点，再写该条在该章所处的顺序，如第4章第1条，用“4．1”表示，后空一格书写该条的标题。如果一章的各条根据需要再

文件资料标准化管理制度

文件资料标准化管理制度 1 目的为进一步加强文件管理，规范文件流程，提高工作效率和办文效率，充分发挥文件上传下达的作用，确保公司政令畅通，特制定本制度。 2 适用范围本制度适用于公司文件资料的管理. 本制度指的文件资料包括公文、管理制度、计划、外来文件和资料等。 3职责权限行政办负责以公司名义上报、下发执行的各类文件资料的编号、发放、收回、归档和销毁。其它各部门负责以部门名义上报、下发的各类文件资料的管理，行政办负责文件资料审查、打印、校对工作。各部门的对外文件，经行政部文字格式的审核，上报主管副总经理审批。 4 术语定义公文——指总公司在处理各种公务时使用的应用型文书，包括：决定、决议、通知、通报、报告、请示、批复、函、会议纪要。管理制度——指要求公司成员共同遵守，按一定程序工作、达到一定标准，并进行考核的文件。计划——指总公司为完成一定时期内的工作任务，而事先做出安排的文件。 5 工作内容及程序文件资料的编写格式 5.1.1 文件资料的用纸标准型纸张。图样表格等不宜减小时，该页必须按以上尺寸折叠装订。一般用型A 4 5.1.2 文件和资料章、条、款的编排规则 5.1.2.1 根据文件内容的编排划分，章就是一个章节，包括条和款，条是章的一个部分，款就是章或条的一个层次。 5.1.2.2 章用阿拉伯数字表示，第一章就用“1”表示，以下按顺序依次类推。“章”应左起空两格书写，右侧空一格写该章的标题，该标题一行书写不够，可另起一行，但这一行的第一个字与该标题的第一个字平排书写。章与章之间应空一行编排。 5.1.2.3 “条”用阿拉伯数字表示，分开章的标题另起一行，先左起空两格，书写该条所处章的编号，后加一个小圆点再写该条在该章所处的顺序，如第“4”章第一条，就用“4.1”表示，右侧空一格书写该条的标题内容。一章的各条根据需要再分为若干下一层次的小条，其编号表示方法同上，如第4章第2条，再分3小条，则分另表示为“4.2.1”、“4.2.2”、“4.2.3”，如果“4.2.1”再分为2个小条，则分别表示为“1.2”;条一般只划到第三层次，即只以4位数字表示为限，下一层次的内容用“款”的形式进行叙述。一章的各条和下一层次再分的条有无标题，原则上应一致，

文件资料标准化管理制度【最新版】

文件资料标准化管理制度 1目的为进一步加强文件管理，规范文件流程，提高工作效率和办文效率，充分发挥文件上传下达的作用，确保公司政令畅通，特制定本制度。 2适用范围 2.1 本制度适用于公司文件资料的管理. 2.2 本制度指的文件资料包括公文、管理制度、计划、外来文件和资料等。 3职责权限 3.1行政办负责以公司名义上报、下发执行的各类文件资料的编号、发放、收回、归档和销毁。 3.2 其它各部门负责以部门名义上报、下发的各类文件资料的管理，行政办负责文件资料审查、打印、校对工作。

3.4各部门的对外文件，经行政部文字格式的审核，上报主管副总经理审批。 4术语定义 4.1 公文--指总公司在处理各种公务时使用的应用型文书，包括:决定、决议、通知、通报、报告、请示、批复、函、会议纪要。 4.2 管理制度--指要求公司成员共同遵守，按一定程序工作、达到一定标准，并进行考核的文件。 4.3 计划--指总公司为完成一定时期内的工作任务，而事先做出安排的文件。 5工作内容及程序 5.1 文件资料的编写格式 5.1.1 文件资料的用纸标准

型纸张。图样表格等不宜减小时，该页必须按以上尺寸折叠装订。一般用型A4 5.1.2 文件和资料章、条、款的编排规则 5.1.2.1 根据文件内容的编排划分，章就是一个章节，包括条和款，条是章的一个部分，款就是章或条的一个层次。 5.1.2.2 章用阿拉伯数字表示，第一章就用“1”表示，以下按顺序依次类推。“章”应左起空两格书写，右侧空一格写该章的标题，该标题一行书写不够，可另起一行，但这一行的第一个字与该标题的第一个字平排书写。章与章之间应空一行编排。 5.1.2.3 “条”用阿拉伯数字表示，分开章的标题另起一行，先左起空两格，书写该条所处章的编号，后加一个小圆点再写该条在该章所处的顺序，如第“4”章第一条，就用“4.1”表示，右侧空一格书写该条的标题内容。一章的各条根据需要再分为若干下一层次的小条，其编号表示方法同上，如第4章第2条，再分3小条，则分另表示为“4.2.1”、