CN102567608A - Performance evaluation experiment analysis method for photo-electric imaging system - Google Patents
Performance evaluation experiment analysis method for photo-electric imaging system Download PDFInfo
- Publication number
- CN102567608A CN102567608A CN2010106109809A CN201010610980A CN102567608A CN 102567608 A CN102567608 A CN 102567608A CN 2010106109809 A CN2010106109809 A CN 2010106109809A CN 201010610980 A CN201010610980 A CN 201010610980A CN 102567608 A CN102567608 A CN 102567608A
- Authority
- CN
- China
- Prior art keywords
- experiment
- factor
- imaging system
- input
- factors
- Prior art date
- Legal status (The legal status 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 status listed.)
- Granted
Links
Images
Landscapes
- Studio Devices (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention relates to a performance evaluation experiment analysis method for a photo-electric imaging system, which includes the following steps: building a photo-electric imaging system performance prediction model; dividing input factor sets and output index sets in the photo-electric imaging system performance prediction model; conducting screening experiment on the input factor sets; dividing input factor levels of the model according to screening experiment results and building a model input factor level graph; selecting orthogonal graphs according to the model input factor level graph to form experiment schemes; invoking the performance prediction model to calculate the output index sets corresponding to each experiment scheme; and performing range analysis and sensitivity analysis by aid of the output index sets and finishing the experiment analysis process. The performance evaluation experiment analysis method for the photo-electric imaging system can compteltely, typically and equilibrium comparably reflect effects of all experiment input factors on experiment indexes with the fewest experiment times, analyzes and determines the optimum level of all the experiment input factors and primary and secondary relations of all the experiment input factors in effects on the output indexes through a range method, and simultaneously analyzes the sensitivity degree of the input factors on the output indexes through a sensitivity analysis method.
Description
Technical field
The present invention relates to a kind of photo electric imaging system Performance Evaluation technical field, a kind of specifically photo electric imaging system Performance Evaluation experiment analytical method.
Background technology
The photo electric imaging system performance prediction model is research and the research aspect of improving photo electric imaging system; Relate to a plurality of specialties such as scene modeling, atmospheric optics, optical engineering, information processing, vision physiological; Photo electric imaging system is influenced image quality mechanism by too many levels; Survey the recognition capability forecast model through setting up photo electric imaging system, realize the assessment of photo electric imaging system performance and analysis of influential factors.
The photo electric imaging system image quality receives outside nature environment, object scene test condition and imaging system self optical lens, charge-coupled device (CCD) and data and sends the multifactorial influences of too many levels such as transmission reception; The complication system of the many index outputs of multifactor input be can be expressed as, qualitative and quantitative confirm to influence photo electric imaging system detection recognition performance influence factor and influence degree are difficult to.
Summary of the invention
The multiple external factor that is subject to that exists to photo electric imaging system in the prior art influences with internal factor; Be difficult to the qualitative and quantitative weak points such as photo electric imaging system detection recognition performance influence factor and influence degree of confirming to influence, the technical matters that the present invention will solve provides a kind of can confirm to influence the influence factor ordering of photo electric imaging system and the photo electric imaging system Performance Evaluation experiment analytical method of influence degree thereof.
For solving the problems of the technologies described above, the technical scheme that the present invention adopts is:
Photo electric imaging system Performance Evaluation experiment analytical method of the present invention may further comprise the steps:
Set up the photo electric imaging system performance prediction model; In performance prediction model, divide input set of factors and output-index collection; The input set of factors is carried out screening experiment; Divide model input factor level according to the screening experiment result, set up model input factor level table; According to model input factor level table, choose orthogonal table, form experimental program; The invocation performance forecast model calculates the corresponding output-index collection of each experimental program; Utilize the output-index collection to carry out range analysis and sensitivity analysis, finish this experimental analysis process.
The process of said division input set of factors and output-index collection is:
According to photo electric imaging system image quality influence factor, will import set of factors and be divided into three aspects: target background characteristic, physical environment characteristic and imaging system characteristic; The output-index collection comprises photo electric imaging system Performance Evaluation index, i.e. acquisition probability and target identification probability.
The target background characteristic comprises: target sizes, target shape, target background contrast; The physical environment characteristic comprises: sun angle, visibility, observed range, ambient light illumination; The imaging system characteristic comprises: focal length, optical aperture, operation wavelength, optical transmittance, picture dot size, field angle, picture dot fill factor, curve factor, picture dot number and quantum efficiency;
The rigid condition of above-mentioned each characterisitic parameter is: confirm a certain universal display device parameter, it is regarded as normal value; Confirm a certain target size and contrast, its attribute is set to normal value; Confirm a certain typical atmospheric condition.
The process of the input set of factors being carried out screening experiment is:
Choose certain factor in the input set of factors that will investigate, fixing other input factors are done the experiment of a plurality of levels, and detection is chosen this factor and whether experimental index had appreciable impact, select foundation as the factor of further experiment;
Through to a plurality of factor experiment screenings; Confirm in the ambient light illumination, physical environment characteristic in the physical environment characteristic optical aperture, focal length, optical transmittance, picture dot fill factor, curve factor and quantum efficiency in observed range, the imaging system characteristic, be specified to field angle, picture dot size value in the picture system performance simultaneously.
The process of setting up model input factor level table is:
Choose certain factor in the input set of factors; Fixing other input factor values; In selected input factor span, calculate output-index; According to selected factor output-index is influenced amplitude and divide this factor level, successively other input factors are carried out aforesaid operations, form input factor level table.
Choosing orthogonal table formation experimental program is: according to input factor level table, select suitable orthogonal test table, form the experimental program table.
The present invention has following beneficial effect and advantage:
1. the inventive method adopts the orthogonal experiment design to carry out photo electric imaging system performance prediction model emulation experiment; Survey on the recognition capability characterization model basis at photo electric imaging system; Divide its input parameter and output-index; According to experimental design thought, adopt orthogonal design design detection recognition capability characterization model emulation experiment, can comprehensive with minimum experiment number, typical, balanced comparable reflection experiment import of the influence of each factor to experimental index.
2. this method adopts range method and Sensitivity Analysis Method to analyze photoelectronic imaging performance prediction model emulation experiment data on the orthogonal experiment design basis; Confirm respectively to import the optimal level of factor and, adopt the sensitivity of Sensitivity Analysis Method analysis input factor simultaneously through the range method analysis output-index to the primary and secondary relation of the influence of output-index.
Description of drawings
Fig. 1 is a photo electric imaging system Performance Evaluation analysis of experiments process flow diagram of the present invention;
Fig. 2 is a photo electric imaging system task estimated performance model synoptic diagram of the present invention.
Embodiment
As shown in Figure 1, photo electric imaging system Performance Evaluation experiment analytical method of the present invention may further comprise the steps:
Set up the photo electric imaging system performance prediction model;
In performance prediction model, divide input set of factors and output-index collection;
The input set of factors is carried out screening experiment;
Divide model input factor level according to the screening experiment result, set up model input factor level table;
According to model input factor level table, choose orthogonal table, form experimental program;
The invocation performance forecast model calculates the corresponding output-index collection of each experimental program;
Utilize the output-index collection to carry out range analysis and sensitivity analysis, finish this experimental analysis process.
This method based target obtains performance model (TTP) and implements.As shown in Figure 2; Be expressed as picture system performance prediction model in the empty frame of image; Comprise imaging system transmission characteristic, human visual system's threshold property, Johnson's criterion; Wherein the imaging system transmission characteristic comprises modulation transfer function of optical system, detector modulation transfer function, information processing modulation transfer function, a plurality of link transmission characteristics of display modulation transfer function model; Human visual system's characteristic comprises human visual system's modulation transfer function and human eye contrast threshold value function; The imaging system transmission characteristic combines human-eye visual characteristic to obtain being expressed as the system contrast threshold function table of picture system performance resolution and sensitivity overall characteristic, utilizes the target background information of importing, and mainly is meant target and physical environment calculation of parameter target apparent contrast of living in thereof; Thereby obtain the equivalent band periodicity of target, calculate detection and the identification probability of human eye this target according to Johnson's criterion.
H is expressed as picture systems communicate characteristic among the figure; Input representes the imaging system input information; Be target and scene parameter of living in thereof, N representes that each link introducing noise item of imaging system disturbs, and T representes to differentiate threshold value by the definite human eye vision of Johnson's criterion; Output DRC is expressed as the detection recognition capability of the target of picture system performance prediction model output, representes with detection probability and identification probability.
The process of said division input set of factors and output-index collection is: according to photo electric imaging system image quality influence factor, will import set of factors and be divided into three aspects: target background characteristic, physical environment characteristic and imaging system characteristic; The output-index collection comprises photo electric imaging system Performance Evaluation index, i.e. acquisition probability and target identification probability;
The target background characteristic comprises: target sizes, target shape, target background contrast; The physical environment characteristic comprises: sun angle, visibility, observed range, ambient light illumination; The imaging system characteristic comprises: focal length, optical aperture, operation wavelength, optical transmittance, picture dot size, field angle, picture dot fill factor, curve factor, picture dot number and quantum efficiency;
Confirm a certain universal display device parameter; It is regarded as normal value; Confirm a certain size and contrast target, its attribute is set to normal value, confirms a certain typical atmospheric condition; The influence that the performance indexes factor of model unifies mainly to receive under the prerequisite detection system parameter and observed range and visibility in objective condition meets and surveys the purpose that distinguishing indexes is passed judgment on.
The process of the input set of factors being carried out screening experiment is: choose certain factor in the input set of factors that will investigate; Fixing other input factors; Do the experiment of a plurality of levels, detect this and choose factor and whether experimental index is had appreciable impact, select foundation as the factor of next step experiment; Through to a plurality of factor experiment screenings; Confirm in the ambient light illumination, physical environment characteristic in the physical environment characteristic optical aperture, focal length, optical transmittance, picture dot fill factor, curve factor and quantum efficiency in observed range, the imaging system characteristic, be specified to field angle in the picture system performance, the best value of picture dot size simultaneously.
The process of setting up model input factor level table is: choose certain factor in the input set of factors, fixing other input factor values are calculated output-index in selected input factor span; According to selected factor output-index is influenced amplitude and divide this factor level; Level choose the overall range that can represent factor, the reflection factor is carried out aforesaid operations to other input factors successively to the homogeneity of the influence of index; Form input factor level table, as shown in table 1.
Present embodiment is taked orthogonal experiment design, and the orthogonal experiment design advantage is that the reasonable permutation and combination through empirical factor realizes under the minimum experiment number, comprehensively with the influence situation of balanced comparable reflection empirical factor to experimental index.Specific practice: according to input factor level table, select suitable orthogonal test table, form the experimental program table, on this basis, carry out the experiment of TTP model, obtain the experimental index under the different tests arrangement., as shown in table 2.
Table 1
Table 2
Range analysis and sensitivity analysis purpose be on orthogonal experiment data basis, analyze each empirical factor to output-index influence ordering and input pointer influence degree to output-index, the range analysis process is:
1, calculate the same horizontal output index of input factor and, averaging obtains average output-index value;
2, ask the minimum and maximum value of on average output level of this input factor, make difference and try to achieve absolute extreme difference;
3, to each input factor extreme difference ordering, obtain the input factor to output-index influence ordering;
The sensitivity analysis process is:
1, choose every factor two levels and form the orthogonal experiment design proposal, making an experiment obtains data;
2, calculate each factor extreme difference, calculate two horizontal extreme difference averages simultaneously;
3, calculate according to extreme difference average and absolute extreme difference
4, calculate the absolute change amount and the relative variation of factor;
5, according to relative extreme difference and factor relative variation, calculate susceptibility.
For guaranteeing the objectivity and the completeness of experiment, under different visibility and ambient lighting, carry out the orthogonal experiment of a plurality of factors.Experimental result reflects that under same atmospheric conditions and observed range focal length and aperture are the principal elements that influences the photo electric imaging system performance.Simultaneously, researchist's this method capable of using is to other photoelectronic imaging input factor experiment Analysis.
Claims (6)
1. photo electric imaging system Performance Evaluation experiment analytical method is characterized in that may further comprise the steps:
Set up the photo electric imaging system performance prediction model;
In performance prediction model, divide input set of factors and output-index collection;
The input set of factors is carried out screening experiment;
Divide model input factor level according to the screening experiment result, set up model input factor level table;
According to model input factor level table, choose orthogonal table, form experimental program;
The invocation performance forecast model calculates the corresponding output-index collection of each experimental program;
Utilize the output-index collection to carry out range analysis and sensitivity analysis, finish this experimental analysis process.
2. by the described photo electric imaging system Performance Evaluation of claim 1 experiment analytical method, it is characterized in that: the process of said division input set of factors and output-index collection is:
According to photo electric imaging system image quality influence factor, will import set of factors and be divided into three aspects: target background characteristic, physical environment characteristic and imaging system characteristic; The output-index collection comprises photo electric imaging system Performance Evaluation index, i.e. acquisition probability and target identification probability.
3. by the described photo electric imaging system Performance Evaluation of claim 2 experiment analytical method, it is characterized in that:
The target background characteristic comprises: target sizes, target shape, target background contrast; The physical environment characteristic comprises: sun angle, visibility, observed range, ambient light illumination; The imaging system characteristic comprises: focal length, optical aperture, operation wavelength, optical transmittance, picture dot size, field angle, picture dot fill factor, curve factor, picture dot number and quantum efficiency;
The rigid condition of above-mentioned each characterisitic parameter is: confirm a certain universal display device parameter, it is regarded as normal value; Confirm a certain target size and contrast, its attribute is set to normal value; Confirm a certain typical atmospheric condition.
4. by the described photo electric imaging system Performance Evaluation of claim 1 experiment analytical method, it is characterized in that: the process of the input set of factors being carried out screening experiment is:
Choose certain factor in the input set of factors that will investigate, fixing other input factors are done the experiment of a plurality of levels, and detection is chosen this factor and whether experimental index had appreciable impact, select foundation as the factor of further experiment;
Through to a plurality of factor experiment screenings; Confirm in the ambient light illumination, physical environment characteristic in the physical environment characteristic optical aperture, focal length, optical transmittance, picture dot fill factor, curve factor and quantum efficiency in observed range, the imaging system characteristic, be specified to field angle, picture dot size value in the picture system performance simultaneously.
5. by the described photo electric imaging system Performance Evaluation of claim 1 experiment analytical method, it is characterized in that: the process of setting up model input factor level table is:
Choose certain factor in the input set of factors; Fixing other input factor values; In selected input factor span, calculate output-index; According to selected factor output-index is influenced amplitude and divide this factor level, successively other input factors are carried out aforesaid operations, form input factor level table.
6. by the described photo electric imaging system Performance Evaluation of claim 1 experiment analytical method, it is characterized in that: choosing orthogonal table formation experimental program is: according to input factor level table, select suitable orthogonal test table, form the experimental program table.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010610980.9A CN102567608B (en) | 2010-12-29 | 2010-12-29 | Performance evaluation experiment analysis method for photo-electric imaging system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010610980.9A CN102567608B (en) | 2010-12-29 | 2010-12-29 | Performance evaluation experiment analysis method for photo-electric imaging system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102567608A true CN102567608A (en) | 2012-07-11 |
| CN102567608B CN102567608B (en) | 2015-04-08 |
Family
ID=46413000
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010610980.9A Active CN102567608B (en) | 2010-12-29 | 2010-12-29 | Performance evaluation experiment analysis method for photo-electric imaging system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102567608B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015164995A1 (en) * | 2014-04-30 | 2015-11-05 | 中国科学院长春光学精密机械与物理研究所 | Optical system misalignment solution-based method for evaluating stability of optical-mechanical structure |
| CN109088124A (en) * | 2018-08-01 | 2018-12-25 | 桑德集团有限公司 | The control strategy of battery liquid cooling system determines method and device |
| CN110989035A (en) * | 2019-12-19 | 2020-04-10 | 中国空间技术研究院 | Optical remote sensing detection performance evaluation method |
| CN112100769A (en) * | 2020-09-08 | 2020-12-18 | 上海三菱电梯有限公司 | Method for constructing elevator component performance model |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101504317A (en) * | 2009-02-27 | 2009-08-12 | 中国人民解放军海军工程大学 | Apparatus for simple detection of infrared imaging system performance parameter |
-
2010
- 2010-12-29 CN CN201010610980.9A patent/CN102567608B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101504317A (en) * | 2009-02-27 | 2009-08-12 | 中国人民解放军海军工程大学 | Apparatus for simple detection of infrared imaging system performance parameter |
Non-Patent Citations (5)
| Title |
|---|
| 常洪花: ""光电图像背景杂波的定量表征及其对成像系统目标获取性能的影响"", 《中国博士学位论文全文数据库信息科技辑》 * |
| 李江等: ""一种复合模式的人脸识别系统设计"", 《国防科技大学学报》 * |
| 韩玉阁等: ""坦克红外辐射特性影响因素的灵敏度分析"", 《红外与激光工程》 * |
| 高岩等: ""电容层析成像系统传感器参数优化——基于正交试验设计法"", 《计算机工程与应用》 * |
| 高稚允等: "《军用光电系统》", 31 December 1996, 北京理工大学出版社 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015164995A1 (en) * | 2014-04-30 | 2015-11-05 | 中国科学院长春光学精密机械与物理研究所 | Optical system misalignment solution-based method for evaluating stability of optical-mechanical structure |
| CN109088124A (en) * | 2018-08-01 | 2018-12-25 | 桑德集团有限公司 | The control strategy of battery liquid cooling system determines method and device |
| CN110989035A (en) * | 2019-12-19 | 2020-04-10 | 中国空间技术研究院 | Optical remote sensing detection performance evaluation method |
| CN110989035B (en) * | 2019-12-19 | 2022-01-11 | 中国空间技术研究院 | Optical remote sensing detection performance evaluation method |
| CN112100769A (en) * | 2020-09-08 | 2020-12-18 | 上海三菱电梯有限公司 | Method for constructing elevator component performance model |
| CN112100769B (en) * | 2020-09-08 | 2022-08-02 | 上海三菱电梯有限公司 | Method for constructing elevator component performance model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102567608B (en) | 2015-04-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ukhwah et al. | Asphalt pavement pothole detection using deep learning method based on YOLO neural network | |
| CN105987805B (en) | A kind of detection method of projection structure photosystem | |
| KR20160096332A (en) | System and method for prediction of environmental pollution | |
| CN107894284B (en) | A kind of infrared camera wave band comparative approach of combination detection efficiency | |
| CN102567608A (en) | Performance evaluation experiment analysis method for photo-electric imaging system | |
| CN106796180B (en) | For diversity and the dynamic lattice of disfigurement discovery | |
| CN113538347B (en) | Image detection method and system based on efficient bidirectional path aggregation attention network | |
| CN110115022B (en) | Image processing system, optical sensor, and learning device | |
| CN106327468A (en) | Curve detection method and device | |
| CN103776532A (en) | Hyperspectral-imager index optimization method based on remote-sensing application | |
| Blagorodnova et al. | GS-TEC: the Gaia spectrophotometry transient events classifier | |
| Faria et al. | The usage of Strömgren photometry in studies of local group dwarf spheroidal galaxies-Application to Draco: a new catalogue of Draco members and a study of the metallicity distribution function and radial gradients | |
| CN104813217B (en) | For the method for the passive single channel imager for being designed to estimate the depth of field | |
| Kim et al. | Utilizing machine learning for detecting flowering in mid-range digital repeat photography | |
| CN117173590A (en) | Water body abnormality monitoring method based on multisource time sequence remote sensing image | |
| Peter et al. | Morphologies of Galaxies in and around a Protocluster at z= 2.300 | |
| Kimura et al. | Single-epoch supernova classification with deep convolutional neural networks | |
| CN107273886A (en) | A kind of food data analysis method based on cloud computing | |
| Silva et al. | Geostatistics and remote sensing methods in the classification of images of areas cultivated with citrus | |
| WO2016005968A2 (en) | System and method for quantifying reflection e.g. when analyzing laminated documents | |
| KR20220020920A (en) | Particle Quantification Device | |
| US8583575B2 (en) | Process and device for representation of a scanning function | |
| CN108154478B (en) | A kind of remote sensing image processing method | |
| KR102209866B1 (en) | Preprocessing Method of Distortion Image for Ground Based Total Cloude Cover Automatic Retrieval | |
| Priyanka et al. | A step towards inter-operable Unmanned Aerial Vehicles (UAV) based phenotyping; A case study demonstrating a rapid, quantitative approach to standardize image acquisition and check quality of acquired images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |