WO2016208142A1 - 画像処理システム、画像処理方法、およびコンピュータ読取可能な記録媒体 - Google Patents
画像処理システム、画像処理方法、およびコンピュータ読取可能な記録媒体 Download PDFInfo
- Publication number
- WO2016208142A1 WO2016208142A1 PCT/JP2016/002813 JP2016002813W WO2016208142A1 WO 2016208142 A1 WO2016208142 A1 WO 2016208142A1 JP 2016002813 W JP2016002813 W JP 2016002813W WO 2016208142 A1 WO2016208142 A1 WO 2016208142A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- gradient
- band
- image processing
- weight
- Prior art date
Links
- 238000003672 processing method Methods 0.000 title claims description 10
- 238000005457 optimization Methods 0.000 claims abstract description 58
- 238000004364 calculation method Methods 0.000 claims abstract description 25
- 238000000034 method Methods 0.000 claims description 22
- 239000011159 matrix material Substances 0.000 claims description 19
- 238000010801 machine learning Methods 0.000 claims description 4
- 230000002194 synthesizing effect Effects 0.000 claims 2
- 239000002131 composite material Substances 0.000 description 18
- 230000006870 function Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 239000003086 colorant Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 239000004615 ingredient Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
- G06T5/92—Dynamic range modification of images or parts thereof based on global image properties
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
- H04N1/387—Composing, repositioning or otherwise geometrically modifying originals
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
- H04N1/46—Colour picture communication systems
- H04N1/56—Processing of colour picture signals
- H04N1/60—Colour correction or control
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/10—Cameras or camera modules comprising electronic image sensors; Control thereof for generating image signals from different wavelengths
- H04N23/11—Cameras or camera modules comprising electronic image sensors; Control thereof for generating image signals from different wavelengths for generating image signals from visible and infrared light wavelengths
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/45—Cameras or camera modules comprising electronic image sensors; Control thereof for generating image signals from two or more image sensors being of different type or operating in different modes, e.g. with a CMOS sensor for moving images in combination with a charge-coupled device [CCD] for still images
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/667—Camera operation mode switching, e.g. between still and video, sport and normal or high- and low-resolution modes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
- H04N25/40—Extracting pixel data from image sensors by controlling scanning circuits, e.g. by modifying the number of pixels sampled or to be sampled
- H04N25/44—Extracting pixel data from image sensors by controlling scanning circuits, e.g. by modifying the number of pixels sampled or to be sampled by partially reading an SSIS array
- H04N25/443—Extracting pixel data from image sensors by controlling scanning circuits, e.g. by modifying the number of pixels sampled or to be sampled by partially reading an SSIS array by reading pixels from selected 2D regions of the array, e.g. for windowing or digital zooming
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
- H04N25/60—Noise processing, e.g. detecting, correcting, reducing or removing noise
- H04N25/68—Noise processing, e.g. detecting, correcting, reducing or removing noise applied to defects
- H04N25/683—Noise processing, e.g. detecting, correcting, reducing or removing noise applied to defects by defect estimation performed on the scene signal, e.g. real time or on the fly detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
- G06T2207/10036—Multispectral image; Hyperspectral image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Definitions
- the present invention relates to an image processing technique, and relates to an image processing system, an image processing method, and a computer-readable recording medium.
- cameras using sensors that are suitable for imaging various target objects are widely used.
- a monitoring camera using a visible light sensor is widely used.
- cameras using non-visible light sensors such as a near infrared camera and a far infrared camera are widely used for nighttime monitoring.
- a near ultraviolet camera is also commercially available.
- devices for imaging wavelengths longer than the visible light wavelength region such as terahertz waves and radio waves are also commercially available as other cameras.
- Patent Document 1 discloses a method of expressing information of an image group of a plurality of bands using a single color image in order to easily analyze an image group acquired from a plurality of sensors.
- a structure tensor is calculated from gradient information of a plurality of band image groups, and an output image is synthesized using the structure tensor. is doing.
- An object of the present invention is to provide an image processing system capable of displaying one image including important information of each band in order to easily analyze an image group acquired from a plurality of sensors.
- An image processing method and a computer-readable recording medium are provided.
- One aspect of the present invention is a weight determination means for determining a band including important information from an image group acquired from a plurality of sensors, and expressing the importance of the band as a weight; Calculating means for calculating an amount calculated based on the gradient of the image in order to constrain the gradient of the output image from the gradient of each image calculated from the group; calculated based on the gradient of the image And an image optimizing unit that synthesizes the output image using the amount.
- an image processing system that can display one image including important information of each band is provided. it can.
- FIG. 1 is a block diagram illustrating a schematic configuration of an image processing system of related technology disclosed in Patent Literature 1.
- FIG. 1 is a block diagram showing a schematic configuration of an image processing system according to an embodiment of the present invention. It is a block diagram which shows schematic structure of the image optimization part used for the image processing system shown in FIG. 3 is a flowchart for explaining the operation of the image processing system shown in FIG. 2.
- FIG. 1 is a block diagram showing an image processing system described in Patent Document 1 and related technology.
- the image processing system described in Patent Document 1 includes an image input unit 10, a gradient constraint calculation unit 21, a gradient calculation unit 22, an image composition unit 23, and an image output unit 30.
- the image input unit 10 receives a plurality of band image groups and a reference color image. Then, the image input unit 10 records the input image in a memory (not shown) or the like.
- the gradient constraint calculation unit 21 calculates a total sum of gradients, called a structure tensor, for each pixel from gradient information of a plurality of band image groups.
- the gradient calculation unit 22 strictly matches the output image and the calculated structure tensor from the structure tensor calculated by the gradient constraint calculation unit 21 and the gradient of the color image to be referenced, and the gradient of the input color image.
- the gradient of the output image is calculated so that the least square error of the gradient of the output image is minimized.
- the image synthesizer 23 synthesizes the output image from the gradient calculated by the gradient calculator 22 using Poisson synthesis.
- Patent Document 1 shown in FIG. 1 when calculating the structure tensor, the structure tensor is simply calculated based on the sum of gradients without considering the characteristics of each band. For this reason, the image processing system of Patent Document 1 shown in FIG. 1 has a problem that important information included in a specific band is not reflected in the output color image.
- the structure tensor is calculated as the sum of the gradient calculated from the visible image and the gradient calculated from the near-infrared image. For this reason, in the image processing system of Patent Document 1 shown in FIG. 1, as a result, the output color image becomes an unclear image due to the influence of the gradient information of the input visible image.
- FIG. 2 is a block diagram showing a schematic configuration of the image processing system according to the embodiment of the present invention.
- an image processing system includes an image input unit 100, a computer (central processing unit; processor; data processing unit) 200 that operates by program control, an image output unit 300, and the like. Consists of.
- the computer (central processing unit; processor; data processing unit) 200 includes a band weight determination unit 210, a gradient constraint calculation unit 220, and an image optimization unit 230. Further, as shown in FIG. 3, the image optimization unit 230 includes a gradient optimization unit 231 and a pixel value optimization unit 232.
- the image obtained by a camera or the like is input to the image input unit 100.
- images to be input color images and images acquired from other sensors may be input separately.
- the image input unit 100 records the input image in a memory (not shown) or the like.
- the red, green, and blue pixel values of the i-th pixel are represented as Ri, Gi, and Bi. Also, these ingredients are put together, It shall be expressed as Furthermore, when there is an image acquired from another sensor other than the input color image, the pixel value of the i-th pixel is also expressed by using a subscript. For example, in the case where a near-infrared image is input in addition to the input color image, the pixel value of the i-th near-infrared image may be expressed as Ni.
- the composite image is represented by a matrix in which the pixel values of each pixel are arranged in the raster scan order as in the input image. More specifically, when an RGB image is considered as an output color image, the red, green, and blue pixel values of the i-th pixel are represented as Ri, Gi, and Bi.
- the image output unit 300 is an output device that outputs a reconstructed image.
- the image output unit 300 is realized by, for example, a display device.
- the band weight determination unit 210 determines a weight indicating which band is important among the input images input by the image input unit 100.
- the importance of each band may be given in advance by the user as a parameter, or may be automatically calculated using a technique such as machine learning. Also, the importance of each band may be the same for the entire image, or a different value may be used for each pixel.
- a multiband image group includes a visible image and a near-infrared image in which a fog region is captured.
- the fog area is unclear and noisy, but in the near-infrared image, there is little noise and a distant landscape is clearly captured.
- the band weight determination unit 210 is set so that the importance of the band of the near-infrared image is increased only in the fog region, and the importance of the band of the visible image (that is, RGB) is large in the other region. What is necessary is just to set.
- a band weight matrix having the i-th band weight determined by the band weight determination unit 210 as the diagonal component of each band weight It shall be expressed using.
- a case where a near infrared image is input in addition to the input color image will be described as an example.
- the weights of the i-th RGB and near-infrared images are determined as W Ri , W Gi , W Bi , and W Ni .
- the i-th band weight matrix Can be expressed as shown in Equation 4 below, for example.
- the gradient constraint calculation unit 220 is a band weight matrix that represents the importance of each band obtained by the input image group input by the image input unit 100 and the band weight determination unit 210.
- the gradient constraint calculation unit 220 includes the i-th gradient matrix. May be calculated as in the following Expression 9.
- ⁇ x is a differential value in the horizontal direction
- ⁇ y is a differential value in the vertical direction.
- the gradient constraint calculation unit 220 uses a band weight matrix that represents the importance of each band. And gradient matrix And the structure tensor May be calculated as shown in Equation 13 below.
- the image optimization unit 230 calculates the structure tensor calculated by the gradient constraint calculation unit 220 based on the value corresponding to the structure tensor calculated from the output composite image.
- the combined image is generated so that the colors of the input color image and the output color image match as much as possible.
- the composite image Represented by Since the composite image has three RGB components for each pixel, Is an N ⁇ 3 matrix when the number of pixels of the composite image is N.
- the image optimization unit 230 specifically calculates, for example, an optimization function represented by the following Expression 18. It is sufficient to generate a composite image by minimizing.
- the first term on the right side of Equation 18 is the structure tensor calculated by the gradient constraint calculation unit 220.
- the value corresponding to the structure tensor calculated from the composite image Is a term that constrains to match.
- the subscript F means the Frobenius norm of the matrix.
- Equation 18 is a term that restricts the colors of the input color image and the synthesized image to match as much as possible.
- Equation 18 The third term on the right side of Equation 18 is a term for matching the pixel value of the input image with the pixel value of the composite image as much as possible.
- the subscript “2” means the L2 norm of the vector.
- ⁇ , ⁇ , and ⁇ are parameters determined in advance by the user.
- the image optimization unit 230 calculates the optimization function expressed by Equation 18.
- Equation 18 are further composed of a gradient optimization unit 231 and a pixel value optimization unit 232.
- the image optimization unit 230 uses the gradient optimization unit 231 and the pixel value optimization unit 232 to perform image gradient. And pixel value Optimize alternately. More specifically, the image optimization unit 230 To optimize about The gradient of Are minimized independently for each pixel.
- the gradient optimization unit 231 extends the optimization function expressed by Expression 18 as shown in Expression 31 below.
- Equation 18 In the equation 31, In addition to Optimize at the same time.
- Equation 31 since the first term, the third term, and the fourth term on the right side of Equation 31 are equivalent to Equation 18, description thereof is omitted.
- Equation 31 The second term on the right side of Equation 31 is When Is a term for constraining to match, and ⁇ is the gradient of the image in the gradient optimization unit 231 and the pixel value optimization unit 232. And pixel value Is a parameter that increases each time the values are alternately optimized, and finally has a very large value.
- the gradient optimization unit 231 calculates, for example, the following formula 46:
- the gradient of the composite image may be calculated by minimizing the.
- the part corresponding to can be calculated independently for each pixel, for example, by using a technique such as parallel calculation Can be calculated at high speed.
- the pixel value of the image can be calculated at high speed by using, for example, Fourier transform.
- the image optimization unit 230 includes the gradient optimization unit 231 and the pixel value optimization unit 232, and in addition to Equation 18, a new auxiliary variable Is added, and an optimization function corresponding to Equation 31 is configured, and a portion (Equation 46) that can be optimized independently for each pixel and image transformation such as Fourier transformation can be efficiently calculated for the entire image.
- Equation 50 the optimization function corresponding to Eq. 18 can be minimized efficiently.
- the image input unit 100 inputs a color image and a multiband image acquired from a plurality of sensors (step S200).
- the band weight determination unit 210 determines the importance of each band (step S201).
- the structure tensor Is calculated (step S202).
- Auxiliary variable corresponding to the gradient of the composite image in the gradient optimization unit 231 Is optimized using, for example, Equation 46 (step S203).
- the pixel value optimization unit 232 the pixel value of the composite image Is optimized using, for example, Formula 50 (step S204).
- the image optimization unit 230 uses auxiliary variables. And the pixel value of the composite image Is increased (step 205).
- the image optimization unit 230 calculates the pixel value It is determined whether the value of is sufficiently converged (step S206). If enough pixel value Is not converged (No in step S206), the image optimization unit 230 repeats the processing from step S203 to S205 again (step S206).
- step S207 If the pixel value Is sufficiently converged (Yes in step S206), the image output unit 300 outputs a composite image composed of the pixel values (step S207).
- a composite image in which important information included in each band is aggregated can be generated from a plurality of images having different properties.
- the reason is that the band weight determination unit 210 determines a weight indicating which of the input images is important, and the gradient constraint calculation unit 220 calculates the structure tensor using the determined band weight. This is because, by calculating, a composite image in which the gradient information of the band is more reflected can be generated.
- a composite image can be generated at higher speed and higher accuracy by alternately optimizing the gradient of the output image and the pixel value of the output image.
- the image optimization unit 230 includes a gradient optimization unit 231 and a pixel value optimization unit 232, configures an optimization function in which auxiliary variables are added to the original optimization function, and further, for each pixel.
- an image processing program is expanded in RAM (random access memory), and hardware such as a control unit (CPU (central processing unit)) is operated based on the program.
- CPU central processing unit
- Each part is realized as various means.
- the program may be recorded on a computer-readable recording medium and distributed.
- the program recorded in the computer-readable recording medium is read into a memory via a wired, wireless, or computer-readable recording medium itself, and operates a control unit and the like.
- Examples of computer-readable recording media include optical disks, magnetic disks, semiconductor memory devices, and hard disks.
- a band weight determination unit 210 a gradient constraint calculation unit 220, and an image optimization unit are based on an image processing program developed in a RAM. It can be realized by operating as 230.
- the embodiment of the present invention it is possible to generate a composite image in which important information included in each band is aggregated from a plurality of images having different properties. Further, according to the embodiment of the present invention, it is possible to generate a composite image with higher speed and higher accuracy by alternately optimizing the gradient of the output image and the pixel value of the output image.
- the specific configuration of the present invention is not limited to the above-described embodiment, and changes in a range not departing from the gist of the present invention are included in the present invention.
- the structure tensor is used as the amount calculated based on the gradient of the image, but the present invention is not limited to this.
- the band weight determination unit determines that the image quality that the image analyst thinks is optimal. Adjustments can be made by determining the weights that represent which bands are important.
- the image quality that the analyst thinks is optimal is used to adjust the weight in the band weight determination unit and analyze the image. Is also applicable.
- the band weight adjusted by each image analyst is used by other image analysts, so that each image analyst can share the experience and know-how of image analysis. It can also be applied to.
- image input unit 200 computer (central processing unit; processor; data processing unit) 210 Band weight determination unit 220 Gradient constraint calculation unit 230 Image optimization unit 231 Gradient optimization unit 232 Pixel value optimization unit 300 Image output unit
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Image Processing (AREA)
Abstract
Description
まず、本発明の理解を容易にするために、上記特許文献1に開示された関連技術の画像処理システムについて説明する。
次に、発明を実施するための形態について図面を参照して詳細に説明する。
図2は、本発明の実施の形態に係る画像処理システムの概略構成を示すブロック図である。
次に、図4のフローチャートを参照して、本実施の形態に係る画像処理システムの全体の動作について詳細に説明する。
次に、本実施の形態の効果について説明する。
200 コンピュータ(中央処理装置;プロセッサ;データ処理装置)
210 バンド重み決定部
220 勾配制約算出部
230 画像最適化部
231 勾配最適化部
232 画素値最適化部
300 画像出力部
Claims (18)
- 複数のセンサより取得された画像群から、重要な情報を含むバンドを決定し、そのバンドの重要度を重みとして表現する重み決定手段と、
前記重みを用いて、前記画像群から算出された各画像の勾配から、出力画像の勾配を制約するために、画像の勾配をもとに算出された量を算出する算出手段と、
前記画像の勾配をもとに算出された量を用いて、前記出力画像を合成する画像最適化手段と、
を含む画像処理システム。 - 前記画像の勾配をもとに算出された量が、構造テンソルから成る、請求項1に記載の画像処理システム。
- 前記画像最適化手段は、
前記出力画像の勾配を画素毎に独立に最適化する勾配最適化手段と、
前記最適化された勾配を用いて画像全体で前記出力画像の画素値を最適化する画素値最適化手段と、
を含む請求項1又は2に記載の画像処理システム。 - 前記重み決定手段は、各バンドの重要度を、パラメータとして予め与えるか、又は機械学習技術を用いて自動的に算出する、請求項1から3の何れか1つに記載の画像処理システム。
- 前記重み決定手段は、各バンドの重要度を画素毎に決定する、請求項1から4の何れか1つに記載の画像処理システム。
- 前記算出手段は、各バンドの重要度を表すバンド重み行列と、各画素の勾配を表す勾配行列とを用いて、前記画像の勾配をもとに算出された量を算出する、請求項1から5の何れか1つに記載の画像処理システム。
- 複数のセンサより取得された画像群を解析して、1枚の画像を得る画像処理システムの画像処理方法であって、
重み決定部が、前記画像群から、重要な情報を含むバンドを決定し、そのバンドの重要度を重みとして表現する重み決定工程と、
算出部が、前記重みを用いて、前記画像群から算出された各画像の勾配から、出力画像の勾配を制約するために、画像の勾配をもとに算出された量を算出する算出工程と、
画像最適化部が、前記画像の勾配をもとに算出された量を用いて、前記出力画像を合成する画像最適化工程と、
を含む画像処理方法。 - 前記画像の勾配をもとに算出された量が、構造テンソルから成る、請求項7に記載の画像処理方法。
- 前記画像最適化工程は、
勾配最適化部が、前記出力画像の勾配を画素毎に独立に最適化する勾配最適化工程と、
画素値最適化部が、前記最適化された勾配を用いて画像全体で前記出力画像の画素値を最適化する画素値最適化工程と、
を含む請求項7又は8に記載の画像処理方法。 - 前記重み決定工程では、前記重み決定部が、各バンドの重要度を、パラメータとして予め与えるか、又は機械学習技術を用いて自動的に算出する、請求項7から9の何れか1つに記載の画像処理方法。
- 前記重み決定工程では、前記重み決定部が各バンドの重要度を画素毎に決定する、請求項7から10の何れか1つに記載の画像処理方法。
- 前記算出工程では、前記算出部が、各バンドの重要度を表すバンド重み行列と、各画素の勾配を表す勾配行列とを用いて、前記画像の勾配をもとに算出された量を算出する、請求項7から11の何れか1つに記載の画像処理方法。
- コンピュータに、複数のセンサより取得された画像群を解析させて、1枚の画像を得させる画像処理システムの画像処理プログラムを記録したコンピュータ読取可能な記録媒体であって、前記画像処理プログラムは、前記コンピュータに、
前記画像群から、重要な情報を含むバンドを決定し、そのバンドの重要度を重みとして表現する重み決定手順と、
前記重みを用いて、前記画像群から算出された各画像の勾配から、出力画像の勾配を制約するために、画像の勾配をもとに算出された量を算出する算出手順と、
前記画像の勾配をもとに算出された量を用いて、前記出力画像を合成する画像最適化手順と、
を実行させるコンピュータ読取可能な記録媒体。 - 前記画像の勾配をもとに算出された量が、構造テンソルから成る、請求項13に記載のコンピュータ読取可能な記録媒体。
- 前記画像最適化手順は、前記コンピュータに、
前記出力画像の勾配を画素毎に独立に最適化する勾配最適化手順と、
前記最適化された勾配を用いて画像全体で前記出力画像の画素値を最適化する画素値最適化手順と、
を実行させる請求項13又は14に記載のコンピュータ読取可能な記録媒体。 - 前記重み決定手順は、前記コンピュータに、各バンドの重要度を、パラメータとして予め与えるか、又は機械学習技術を用いて自動的に算出させる、請求項13から15の何れか1つに記載のコンピュータ読取可能な記録媒体。
- 前記重み決定手順は、前記コンピュータに、各バンドの重要度を画素毎に決定させる、請求項13から16の何れか1つに記載のコンピュータ読取可能な記録媒体。
- 前記算出手順は、前記コンピュータに、各バンドの重要度を表すバンド重み行列と、各画素の勾配を表す勾配行列とを用いて、前記画像の勾配をもとに算出された量を算出させる、請求項13から17の何れか1つに記載のコンピュータ読取可能な記録媒体。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US15/738,189 US11010876B2 (en) | 2015-06-26 | 2016-06-10 | Image processing system, image processing method, and computer-readable recording medium |
JP2017524597A JP6826277B2 (ja) | 2015-06-26 | 2016-06-10 | 画像処理システム、画像処理方法、および画像処理プログラム |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2015129122 | 2015-06-26 | ||
JP2015-129122 | 2015-06-26 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2016208142A1 true WO2016208142A1 (ja) | 2016-12-29 |
Family
ID=57585522
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2016/002813 WO2016208142A1 (ja) | 2015-06-26 | 2016-06-10 | 画像処理システム、画像処理方法、およびコンピュータ読取可能な記録媒体 |
Country Status (3)
Country | Link |
---|---|
US (1) | US11010876B2 (ja) |
JP (1) | JP6826277B2 (ja) |
WO (1) | WO2016208142A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110706188A (zh) * | 2019-09-23 | 2020-01-17 | 北京航天宏图信息技术股份有限公司 | 一种影像融合方法、装置、电子设备及存储介质 |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109656107B (zh) * | 2019-01-30 | 2020-05-26 | 北京理工大学 | 一种采用变换l1范数约束的光刻光源图形优化方法 |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2009224971A (ja) * | 2008-03-14 | 2009-10-01 | Omron Corp | 画像処理装置 |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6778698B1 (en) * | 1999-06-11 | 2004-08-17 | Pts Corporation | Method and apparatus for digital image segmentation |
EP1530156B1 (en) * | 2003-11-07 | 2012-03-14 | Mitsubishi Electric Information Technology Centre Europe B.V. | Visual object detection |
US8233705B2 (en) * | 2008-10-03 | 2012-07-31 | Eastman Kodak Company | Potential field-based gamut mapping |
JP2010103740A (ja) | 2008-10-23 | 2010-05-06 | Panasonic Corp | デジタルカメラ |
GB0914982D0 (en) | 2009-08-27 | 2009-09-30 | Univ East Anglia | Methods and apparatus for generating accented image data |
US8903130B1 (en) * | 2011-05-09 | 2014-12-02 | Google Inc. | Virtual camera operator |
US8891866B2 (en) * | 2011-08-31 | 2014-11-18 | Sony Corporation | Image processing apparatus, image processing method, and program |
CN104700381B (zh) * | 2015-03-13 | 2018-10-12 | 中国电子科技集团公司第二十八研究所 | 一种基于显著目标的红外与可见光图像融合方法 |
-
2016
- 2016-06-10 US US15/738,189 patent/US11010876B2/en active Active
- 2016-06-10 WO PCT/JP2016/002813 patent/WO2016208142A1/ja active Application Filing
- 2016-06-10 JP JP2017524597A patent/JP6826277B2/ja active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2009224971A (ja) * | 2008-03-14 | 2009-10-01 | Omron Corp | 画像処理装置 |
Non-Patent Citations (3)
Title |
---|
SHIN YOSHIZAWA ET AL.: "Seamless Cloning with Texture Restoration via Poisson Image Analogy", IPSJ SIG NOTES COMPUTER GRAPHICS AND CAD(CG, vol. 2014 -CG, no. 6, 13 February 2014 (2014-02-13), pages 1 - 10 * |
TAKASHI SHIBATA ET AL.: "Kashi Gazo to Kinsekigai Gazo no Togo ni yoru Gazo no Shininsei Kojo", SSII2015 DAI 21 KAI SYMPOSIUM ON SENSING VIA IMAGE INFORMATION KOEN RONBUNSHU, 10 June 2015 (2015-06-10), pages 1 - 6 * |
YOSUKE NAGATOMO ET AL.: "A Fast Poisson Image Synthesis Algorithm using Image Subsampling and Gradient-based Interpolation", IEICE TECHNICAL REPORT, vol. 114, no. 172, 25 July 2014 (2014-07-25), pages 47 - 52 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110706188A (zh) * | 2019-09-23 | 2020-01-17 | 北京航天宏图信息技术股份有限公司 | 一种影像融合方法、装置、电子设备及存储介质 |
Also Published As
Publication number | Publication date |
---|---|
US20180174282A1 (en) | 2018-06-21 |
JPWO2016208142A1 (ja) | 2018-04-12 |
JP6826277B2 (ja) | 2021-02-03 |
US11010876B2 (en) | 2021-05-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Maurer | How to pan-sharpen images using the gram-schmidt pan-sharpen method–A recipe | |
JP4767718B2 (ja) | 画像処理方法および装置ならびにプログラム | |
JP2019534520A (ja) | 画像処理用のニューラルネットワークモデルのトレーニング方法、装置、及び記憶媒体 | |
US20060269270A1 (en) | Photography apparatus, photography method and photography program | |
JP7328096B2 (ja) | 画像処理装置、画像処理方法、およびプログラム | |
JP2012239103A (ja) | 画像処理装置、および画像処理方法、並びにプログラム | |
JP5546166B2 (ja) | 撮像装置、信号処理方法、及びプログラム | |
US20150332434A1 (en) | Demosaicking System and Method for Color array Based Multi-Spectral Sensors | |
JP2010062672A (ja) | 画像処理装置およびその方法 | |
WO2021106853A1 (ja) | 画像処理装置、画像処理方法、学習装置、学習方法、撮像装置、及びプログラム | |
JP6909979B2 (ja) | 画像処理システム、画像処理方法、および画像処理プログラム | |
JP2018036960A (ja) | 画像類似度算出装置、画像処理装置、画像処理方法、及び記録媒体 | |
JP7370922B2 (ja) | 学習方法、プログラム及び画像処理装置 | |
JP2020061080A (ja) | 画像処理装置、撮像装置、画像処理方法 | |
WO2016208142A1 (ja) | 画像処理システム、画像処理方法、およびコンピュータ読取可能な記録媒体 | |
US11543644B2 (en) | Digital imaging device and method for generating a digital color image | |
JP4615430B2 (ja) | 画像生成装置、画像生成方法および画像生成プログラム | |
JP6835227B2 (ja) | 画像処理装置、画像処理方法およびコンピュータプログラム | |
JP5846885B2 (ja) | 画像処理装置及び画像処理方法 | |
JP6230504B2 (ja) | 画像処理装置、方法、及びプログラム | |
JP2011043901A (ja) | 画像処理装置、画像処理方法、画像処理プログラム、および、電子機器 | |
JP2011095110A (ja) | 光源推定装置および光源推定方法 | |
US20230088317A1 (en) | Information processing apparatus, information processing method, and storage medium | |
US10943561B2 (en) | Image data display system, image data display method, and image data display program recording medium | |
Barzel et al. | SEL-CIE: Knowledge-Guided Self-Supervised Learning Framework for CIE-XYZ Reconstruction from Non-Linear sRGB Images |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16813916 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2017524597 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 15738189 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 16813916 Country of ref document: EP Kind code of ref document: A1 |