JPWO2023090152A5 - - Google Patents
Download PDFInfo
- Publication number
- JPWO2023090152A5 JPWO2023090152A5 JP2023561514A JP2023561514A JPWO2023090152A5 JP WO2023090152 A5 JPWO2023090152 A5 JP WO2023090152A5 JP 2023561514 A JP2023561514 A JP 2023561514A JP 2023561514 A JP2023561514 A JP 2023561514A JP WO2023090152 A5 JPWO2023090152 A5 JP WO2023090152A5
- Authority
- JP
- Japan
- Prior art keywords
- optical
- parameter
- image processing
- optical system
- unit
- 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.)
- Pending
Links
- 230000003287 optical effect Effects 0.000 claims description 58
- 238000011156 evaluation Methods 0.000 claims description 5
- 238000003672 processing method Methods 0.000 claims description 2
Description
本開示にかかる画像処理装置は、対象物の撮影を行う光学系と、オペレータが光学系の光学パラメータを複数入力する設定入力部と、設定入力部に入力された複数の光学パラメータを記憶する光学パラメータ記憶部と、光学パラメータ記憶部に記憶した複数の光学パラメータを順に選択し、選択した光学パラメータに基づいて光学系を配置する光学系制御部と、光学系制御部により配置された光学系が出力した複数の対象物の画像より各々特徴量を算出する画像処理部と、画像処理部により算出された複数の特徴量より目的関数を各々算出する評価部と、評価部が算出した目的関数より、多項式またはガウス関数である近似関数を算出する近似関数作成部と、目的関数が最小となる光学パラメータを基準パラメータセットとし、基準パラメータセットにおける近似関数作成部により出力された近似関数の勾配を算出し、勾配が最大となる方向の近傍の新たな追加光学パラメータを新たな光学パラメータとして光学パラメータ記憶部へ追加する探索部とを備えたものである。 An image processing device according to the present disclosure includes an optical system that photographs an object, a setting input section through which an operator inputs a plurality of optical parameters of the optical system, and an optical system that stores the plurality of optical parameters input into the setting input section. a parameter storage unit; an optical system control unit that sequentially selects a plurality of optical parameters stored in the optical parameter storage unit; and an optical system control unit that arranges an optical system based on the selected optical parameters; An image processing unit that calculates feature quantities from the output images of a plurality of objects, an evaluation unit that calculates objective functions from the plurality of feature quantities calculated by the image processing unit, and an objective function calculated by the evaluation unit. , an approximation function generation unit that calculates an approximation function that is a polynomial or a Gaussian function , and an optical parameter that minimizes the objective function as a reference parameter set, and calculates the gradient of the approximation function output by the approximation function generation unit in the reference parameter set. and a search unit that adds a new additional optical parameter near the direction in which the gradient is maximum to the optical parameter storage unit as a new optical parameter.
また、本開示にかかる画像処理方法は、光学系を用いて対象物の撮影を行うステップと、光学系の光学パラメータを複数入力するステップと、入力された複数の前記光学パラメータを記憶するステップと、記憶した複数の前記光学パラメータを順に選択し、選択した前記光学パラメータに基づいて前記光学系を配置するステップと、配置された前記光学系が出力した複数の前記対象物の画像より各々特徴量を算出するステップと、算出された複数の前記特徴量より目的関数を各々算出するステップと、算出した前記目的関数より、多項式またはガウス関数である近似関数を算出するステップと、前記目的関数が最小となる前記光学パラメータを基準パラメータセットとし、前記基準パラメータセットにおける前記近似関数の勾配を算出し、前記勾配が最大となる方向の近傍の新たな追加光学パラメータを新たな前記光学パラメータとして記憶するステップとを備えたものである。
Further, the image processing method according to the present disclosure includes the steps of photographing an object using an optical system, inputting a plurality of optical parameters of the optical system, and storing the inputted plurality of optical parameters. , sequentially selecting the plurality of stored optical parameters and arranging the optical system based on the selected optical parameters; and determining feature quantities from each of the plurality of images of the object outputted by the arranged optical system. a step of calculating an objective function from each of the plurality of calculated feature quantities; a step of calculating an approximation function that is a polynomial or a Gaussian function from the calculated objective function; A step of setting the optical parameters that become a reference parameter set, calculating the gradient of the approximation function in the reference parameter set, and storing a new additional optical parameter near the direction in which the gradient is maximum as the new optical parameter. It is equipped with the following.
Claims (7)
オペレータが前記光学系の光学パラメータを複数入力する設定入力部と、
前記設定入力部に入力された複数の前記光学パラメータを記憶する光学パラメータ記憶部と、
前記光学パラメータ記憶部に記憶した複数の前記光学パラメータを順に選択し、選択した前記光学パラメータに基づいて前記光学系を配置する光学系制御部と、
前記光学系制御部により配置された前記光学系が出力した複数の前記対象物の画像より各々特徴量を算出する画像処理部と、
前記画像処理部により算出された複数の前記特徴量より目的関数を各々算出する評価部と、
前記評価部が算出した前記目的関数より、多項式またはガウス関数である近似関数を算出する近似関数作成部と、
前記目的関数が最小となる前記光学パラメータを基準パラメータセットとし、前記基準パラメータセットにおける前記近似関数作成部により出力された前記近似関数の勾配を算出し、前記勾配が最大となる方向の近傍の新たな追加光学パラメータを新たな前記光学パラメータとして前記光学パラメータ記憶部へ追加する探索部と、
を備える画像処理装置。 an optical system for photographing the object;
a setting input section through which an operator inputs a plurality of optical parameters of the optical system;
an optical parameter storage unit that stores the plurality of optical parameters input to the setting input unit;
an optical system control unit that sequentially selects the plurality of optical parameters stored in the optical parameter storage unit and arranges the optical system based on the selected optical parameters;
an image processing unit that calculates a feature quantity from each of the plurality of images of the object output by the optical system arranged by the optical system control unit;
an evaluation unit that calculates an objective function from each of the plurality of feature quantities calculated by the image processing unit;
an approximation function creation unit that calculates an approximation function that is a polynomial or a Gaussian function from the objective function calculated by the evaluation unit;
The optical parameters that minimize the objective function are set as a reference parameter set, the gradient of the approximation function output by the approximation function creation unit in the reference parameter set is calculated, and a new value in the vicinity in the direction where the gradient is maximum is calculated. a search unit that adds an additional optical parameter to the optical parameter storage unit as a new optical parameter;
An image processing device comprising:
前記光学系の光学パラメータを複数入力するステップと、
入力された複数の前記光学パラメータを記憶するステップと、
記憶した複数の前記光学パラメータを順に選択し、選択した前記光学パラメータに基づいて前記光学系を配置するステップと、
配置された前記光学系が出力した複数の前記対象物の画像より各々特徴量を算出するステップと、
算出された複数の前記特徴量より目的関数を各々算出するステップと、
算出した前記目的関数より、多項式またはガウス関数である近似関数を算出するステップと、
前記目的関数が最小となる前記光学パラメータを基準パラメータセットとし、前記基準パラメータセットにおける前記近似関数の勾配を算出し、前記勾配が最大となる方向の近傍の新たな追加光学パラメータを新たな前記光学パラメータとして記憶するステップと、
を備える画像処理方法。 a step of photographing the object using an optical system;
inputting a plurality of optical parameters of the optical system;
storing the plurality of input optical parameters;
sequentially selecting the plurality of stored optical parameters and arranging the optical system based on the selected optical parameters;
calculating a feature amount from each of the plurality of images of the object outputted by the arranged optical system;
calculating an objective function from each of the plurality of calculated feature quantities;
calculating an approximate function that is a polynomial or a Gaussian function from the calculated objective function;
The optical parameter where the objective function is the minimum is set as a reference parameter set, the gradient of the approximation function in the reference parameter set is calculated, and a new additional optical parameter near the direction where the gradient is maximum is set as a new optical parameter. a step of memorizing it as a parameter;
An image processing method comprising:
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2021186736 | 2021-11-17 | ||
PCT/JP2022/040899 WO2023090152A1 (en) | 2021-11-17 | 2022-11-01 | Image processing device and image processing method |
Publications (2)
Publication Number | Publication Date |
---|---|
JPWO2023090152A1 JPWO2023090152A1 (en) | 2023-05-25 |
JPWO2023090152A5 true JPWO2023090152A5 (en) | 2024-04-01 |
Family
ID=86396811
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP2023561514A Pending JPWO2023090152A1 (en) | 2021-11-17 | 2022-11-01 |
Country Status (2)
Country | Link |
---|---|
JP (1) | JPWO2023090152A1 (en) |
WO (1) | WO2023090152A1 (en) |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003270163A (en) * | 2002-03-15 | 2003-09-25 | Dainippon Printing Co Ltd | Inspection method and device |
JP6977634B2 (en) * | 2018-03-13 | 2021-12-08 | オムロン株式会社 | Visual inspection equipment, visual inspection methods and programs |
US10520424B2 (en) * | 2018-04-03 | 2019-12-31 | Hiwin Technologies Corp. | Adaptive method for a light source for inspecting an article |
JP2021092439A (en) * | 2019-12-10 | 2021-06-17 | 池上通信機株式会社 | Illumination optimization method, control device, and program |
-
2022
- 2022-11-01 WO PCT/JP2022/040899 patent/WO2023090152A1/en active Application Filing
- 2022-11-01 JP JP2023561514A patent/JPWO2023090152A1/ja active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
TW202009805A (en) | Image processing method and apparatus | |
KR102349920B1 (en) | Learning method and learning device for object detector based on cnn to be used for multi-camera or surround view monitoring using image concatenation and target object merging network, and testing method and testing device using the same | |
CN109685097B (en) | Image detection method and device based on GAN | |
US20130223759A1 (en) | Image processing method and device, and program | |
CN101795361A (en) | Two-dimensional polynomial model for depth estimation based on two-picture matching | |
CN109829875A (en) | Method and apparatus for estimating disparity | |
JP2014511508A5 (en) | ||
US8817075B2 (en) | Imaging apparatus capable of generating three-dimensional images, three-dimensional image generating method, and recording medium | |
KR102569169B1 (en) | Method and apparatus for automatic generation of artificial neural networks | |
KR101843516B1 (en) | Method for determination of focal length for a zoom lens | |
JP6447840B2 (en) | Image device, method for automatic focusing in an image device, and corresponding computer program | |
JP2020041290A (en) | Optimal compaction determination construction system for concrete | |
JP3333721B2 (en) | Area detection device | |
JP2011188318A5 (en) | ||
JP2018022247A5 (en) | ||
JPWO2023090152A5 (en) | ||
CN114127736A (en) | Apparatus and computer-implemented method for processing digital sensor data and training method therefor | |
CN112734759A (en) | Method and device for determining trigger point of flying shooting | |
CN116385369A (en) | Depth image quality evaluation method and device, electronic equipment and storage medium | |
JP2011047739A (en) | Calibration device for camera parameter and program | |
JP2021135977A (en) | Apparatus and method for processing information | |
CN115022549B (en) | Shooting composition method, shooting composition device, computer equipment and storage medium | |
JP2010187130A (en) | Camera calibrating device, camera calibration method, camera calibration program, and recording medium having the program recorded therein | |
JP2016110622A5 (en) | Control device, lens device, image processing system, control method, image processing method, program, and storage medium | |
KR102440073B1 (en) | Apparatus and method for selecting optimized models |