WO2019041245A1 - 斑点检测方法、系统、存储介质及计算机程序产品 - Google Patents

斑点检测方法、系统、存储介质及计算机程序产品 Download PDF

Info

Publication number
WO2019041245A1
WO2019041245A1 PCT/CN2017/099978 CN2017099978W WO2019041245A1 WO 2019041245 A1 WO2019041245 A1 WO 2019041245A1 CN 2017099978 W CN2017099978 W CN 2017099978W WO 2019041245 A1 WO2019041245 A1 WO 2019041245A1
Authority
WO
WIPO (PCT)
Prior art keywords
pixel
image
processed
pixels
speckle
Prior art date
Application number
PCT/CN2017/099978
Other languages
English (en)
French (fr)
Inventor
冯华亮
杜劼熹
Original Assignee
深圳市大疆创新科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳市大疆创新科技有限公司 filed Critical 深圳市大疆创新科技有限公司
Priority to CN201780004458.6A priority Critical patent/CN108496202A/zh
Priority to PCT/CN2017/099978 priority patent/WO2019041245A1/zh
Publication of WO2019041245A1 publication Critical patent/WO2019041245A1/zh
Priority to US16/803,022 priority patent/US20200202510A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformation in the plane of the image
    • G06T3/40—Scaling the whole image or part thereof
    • G06T3/4023—Decimation- or insertion-based scaling, e.g. pixel or line decimation
    • G06T3/04—
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/98—Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns
    • G06V10/993—Evaluation of the quality of the acquired pattern
    • 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
    • G06T2207/20016—Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
    • 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
    • G06T2207/20021—Dividing image into blocks, subimages or windows
    • 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
    • G06T2207/20024—Filtering details
    • 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
    • G06T2207/20112—Image segmentation details
    • G06T2207/20164—Salient point detection; Corner detection

Definitions

  • the present application relates to the field of image processing, and more particularly to a spot detection method, system, storage medium, and computer program product.
  • the traditional image spot detection method performs spot detection on the image to be processed by traversing each pixel in the image to be processed, but the size of the image to be processed is generally large, and direct detection of the image to be processed may result in low efficiency of image spot detection.
  • the present application provides a speckle detection method, system, storage medium, and computer program product, which may improve the efficiency of image spot detection.
  • a method for detecting a spot includes: acquiring a thumbnail image of an image to be processed; performing spot detection on the thumbnail image to obtain a spot pixel of the thumbnail image; and displaying a spot according to the thumbnail image a pixel that determines a speckle pixel of the image to be processed.
  • an image processing system comprising: a memory for storing a program; a processor for executing a program stored in the memory, wherein when the program is executed, the processor is configured to execute The method described on the one hand.
  • a speckle detection apparatus comprising means for performing the method of the first aspect.
  • a computer readable storage medium comprising computer instructions that, when executed on a computer, cause the computer to perform the method of the first aspect.
  • a computer program product that, when executed on a computer, causes the computer to perform the method of the first aspect.
  • the size of the thumbnail image is small, and the efficiency of the spot detection process of the thumbnail image is relatively high. Further, the spot detection result of the thumbnail image is used to determine the spot pixel in the image to be processed, and the spot detection is performed on the premise that the approximate image of the spot pixel of the image to be processed is obtained, and the whole may be improved. The efficiency of spot detection of images is processed.
  • FIG. 1 is a schematic flowchart of a method for detecting a spot according to an embodiment of the present invention.
  • step 130 of FIG. 2 is a schematic flow diagram of one implementation of step 130 of FIG.
  • FIG. 3 is a schematic flow chart of another implementation of step 130 in FIG. 1.
  • FIG. 4 is a diagram showing an example of an implementation in which a reference pixel of a first pixel is selected based on a distance relationship between pixels.
  • FIG. 5 is a diagram showing an example of an implementation in which a reference pixel of a first pixel is selected based on an adjacent relationship between pixels.
  • FIG. 6 is a schematic flow chart of an implementation of step 320 in FIG.
  • FIG. 7 is a schematic structural diagram of an image processing system according to an embodiment of the present invention.
  • the present application is applicable to various image processing systems that require speckle detection and/or speckle filtering of an image.
  • the image processing system may be, for example, a drone with visual perception, a robot, a driverless car, a submersible, or the like.
  • Embodiments of the present invention provide a speckle detection method, which can improve the speckle detection efficiency of an image.
  • the embodiments of the present invention are described in detail below with reference to FIG. 1.
  • FIG. 1 is a schematic flowchart of a method for detecting a spot according to an embodiment of the present invention.
  • the method of Figure 1 can be performed by the image processing system described above.
  • the processing unit may be, for example, at least one of the following processing units: a central processing unit (CPU), a digital signal processor (DSP) or a hardened IP core (intellictual property core).
  • the method of FIG. 1 may include steps 110-130, which are described in detail below.
  • a thumbnail image of the image to be processed is acquired.
  • the thumbnail image of the image to be processed is also A small image that can be called an image to be processed.
  • the thumbnail image and the object depicted by the image to be processed are the same object.
  • a thumbnail image and a to-be-processed image can describe the same scene or the same person.
  • the thumbnail image differs from the image to be processed in that the resolution of the thumbnail image is smaller than the resolution of the image to be processed. For example, the resolution of the thumbnail image is 256 ⁇ 256, and the resolution of the image to be processed is 1024 ⁇ 1024.
  • the manner of acquiring the thumbnail image is not specifically limited in the embodiment of the present invention.
  • the image to be processed can be downsampled to obtain a thumbnail image.
  • the image to be processed may be sampled to obtain a thumbnail image.
  • the thumbnail image can be a downsampled image or a sampled image of the image to be processed.
  • the pixels of the sampled image may be composed of a partial pixel of the image to be processed (i.e., a pixel as a sampling point in the image to be processed).
  • the thumbnail image is subjected to spot detection to obtain a spot pixel of the thumbnail image.
  • the method for detecting the blob of the thumbnail image is not specifically limited, and the spot detection may be performed on the thumbnail image by using any speckle detection algorithm.
  • the pixel values of each pixel in the thumbnail image may be detected first. If the pixel values of the pixels having the connected relationship in the thumbnail image are close to each other (for example, the difference between the pixel values is smaller than the preset difference), the region in which the pixels having the adjacent relationship are located may be referred to as a connected domain. Second, the area of each connected domain in the thumbnail image can be judged. If the area of the connected domain is less than a preset threshold, the pixels in the connected domain are regarded as speckle pixels.
  • a speckle pixel of a thumbnail image may refer to a pixel in a speckle region in the thumbnail image.
  • the speckle pixel of the thumbnail image may refer to a pixel in the thumbnail image that is determined to be speckle or speckle noise.
  • a speckle pixel of the image to be processed is determined based on the speckle pixel of the thumbnail image.
  • pixels in the image to be processed corresponding to the spot pixels in the thumbnail image may be directly determined as the spot pixels of the image to be processed.
  • the speckle pixel of the image to be processed may be determined according to the pixel value of the pixel corresponding to the speckle pixel of the thumbnail image and the surrounding pixels in the image to be processed. The detailed description will be made in conjunction with specific embodiments.
  • the size of the thumbnail image is small, and the efficiency of the spot detection process of the thumbnail image is relatively high. Further, the spot detection result of the thumbnail image is used to determine the spot pixel in the image to be processed, and the spot detection is performed on the premise that the approximate image of the spot pixel of the image to be processed is obtained, and the whole may be improved. The efficiency of spot detection of images is processed.
  • an image to be processed is required.
  • the data of the entire image is randomly accessed because the shape of the connected domain of the image is uncertain and can be caused to expand in any direction of the image.
  • the image processing system can efficiently and randomly access image data in memory.
  • the image processing system (especially the on-chip image processing system) generally has a small memory capacity and cannot accommodate the entire image data at one time, resulting in the image processing system needing to store the entire image in an external memory and in the external memory.
  • Image data is randomly accessed.
  • the image processing system has low efficiency of random access to the external memory, resulting in a low efficiency of the conventional image spot detection method.
  • the size of the image data of the thumbnail image may be set to be smaller than the capacity of the memory (the capacity of the memory may refer to the total capacity of the memory, and may also refer to the capacity of the current free space of the memory).
  • the method of FIG. 1 may further include reading image data of the thumbnail image from the external memory into the memory (one-time read into the memory); step 120 may include: performing the thumbnail image in the memory Spot detection. Assuming that the ratio of sampling (or downsampling) of the thumbnail image is 1/M, the memory space required for spot detection of the thumbnail image is 1/(M*M) of the memory space required for spot detection of the image to be processed directly. Therefore, the thumbnail image has a lower memory capacity requirement than the image to be processed. Since the thumbnail image is small, it can be read into the memory at one time, and the image processing system does not need to randomly access the external storage device, which will greatly improve the image spot detection efficiency.
  • the size of the image data of the thumbnail image may be set to be smaller than the capacity of the memory (the capacity of the memory may refer to the total capacity of the memory, and may also refer to the capacity of the current free space of the memory).
  • Step 110 may include reading image data of the thumbnail image from the external memory into the memory at one time. Assuming that the ratio of sampling (or downsampling) of the thumbnail image is 1/M, the memory space required for spot detection of the thumbnail image is 1/(M*M) of the memory space required for spot detection of the image to be processed directly. ). Further, in the embodiment of the present invention, since the thumbnail image is small, it can be read into the memory at one time, and the image processing system does not need to randomly access the external storage device, which will greatly improve the spot detection efficiency of the image.
  • step 130 is described in detail below with reference to FIG. 2-6.
  • the pixels of the thumbnail image are partial pixels of the image to be processed, that is, the pixels of the thumbnail image may be composed of partial pixels of the image to be processed.
  • the thumbnail image can be a sampled image of the image to be processed.
  • the method of FIG. 2 may include steps 210-220, which are described in detail below.
  • a target pixel set is selected from the image to be processed according to the pixel of the thumbnail image.
  • step 220 the pixels in the target pixel set are determined as the spot pixels of the image to be processed.
  • the pixels in the target pixel set may be pixels having a correspondence relationship with the spot pixels of the thumbnail image.
  • the correspondence may mean that after the thumbnail image is enlarged to be equal to the size of the image to be processed, the relative position of the pixel in the target pixel set in the image to be processed is the same as the relative position of the spot pixel in the thumbnail image.
  • the thumbnail image is composed of partial pixels in the image to be processed, and the pixels in the target pixel set may be the same pixels as the spot pixels of the thumbnail image.
  • the pixel corresponding to the spot pixel of the thumbnail image in the image to be processed is directly determined as the spot pixel of the image to be processed.
  • the image to be processed contains pixels ⁇ p0, p1, p2, p3, p4, p5, p6, p7, p8 ⁇
  • the thumbnail image contains pixels ⁇ p0, p3, p6 ⁇
  • the thumbnail is abbreviated.
  • the pixels ⁇ p3, p6 ⁇ in the image are determined as speckle pixels
  • the pixels ⁇ p3, p6 ⁇ of the image to be processed constitute the target pixel set of the image to be processed, and in the implementation corresponding to FIG. 2, the image to be processed
  • the pixels ⁇ p3, p6 ⁇ are directly determined as the spot pixels of the image to be processed.
  • the pixels in the thumbnail image are composed of partial pixels of the image to be processed, and if a certain pixel in the thumbnail image is determined to be a spot pixel, the same probability of the pixel in the image to be processed is also a spot pixel.
  • the corresponding implementation of FIG. 2 fully utilizes the pixel relationship between the thumbnail image and the image to be processed, and directly determines the pixel in the image to be processed that is the same as the pixel in the thumbnail image as the spot pixel of the image to be processed, which simplifies the pending processing.
  • the method of spot detection of images is used to determines the pixel in the image to be processed that is the same as the pixel in the thumbnail image as the spot pixel of the image to be processed, which simplifies the pending processing.
  • FIG. 3 is a schematic flow chart of another implementation of step 130 in FIG. 1.
  • the pixels of the thumbnail image are partial pixels of the image to be processed, that is, the pixels of the thumbnail image may be composed of partial pixels of the image to be processed.
  • the thumbnail image can be a sampled image of the image to be processed.
  • the method of FIG. 3 may include steps 310-320, which are described in detail below.
  • the first pixel in FIG. 3 may be a target pixel set in the image to be processed (the definition of the target pixel set is the same as the definition of the target pixel set in FIG. 2, and may be specifically referred to the related description of FIG. 2). Any one of the remaining pixels.
  • the pixel set formed by the remaining pixels other than the target pixel set in the image to be processed is hereinafter referred to as the second pixel set of the image to be processed, and then the first pixel is the second pixel set of the image to be processed. Any one of the pixels.
  • each of the second set of pixels of the image to be processed may be executed as the first pixel once as shown in FIG.
  • only a part of the pixels in the second set of pixels of the image to be processed may be used as the first pixel to perform the flow as shown in FIG.
  • the user does not care whether or not speckle pixels appear in the edge regions of the image to be processed. Therefore, in this application, the flow shown in FIG. 3 can be performed without using the pixel located in the edge region of the image to be processed in the second pixel set of the image to be processed as the first pixel.
  • a reference pixel of the first pixel is selected from a portion of the pixels of the image to be processed.
  • the reference pixel may be a pixel adjacent to the first pixel of the partial pixels of the image to be processed, or one or more pixels of the partial pixels of the image to be processed that are closest to the first pixel.
  • part of the pixels of the image to be processed in step 310 refers to pixels in the image to be processed for forming a thumbnail image.
  • a part of pixels of an image to be processed is hereinafter referred to as a first type of pixel of an image to be processed, and a remaining pixel other than the first type of pixels in the image to be processed is referred to as a second type of image to be processed. Pixel.
  • a reference pixel of the first pixel may be selected based on a distance relationship between pixels in the image to be processed.
  • the reference pixels of the first pixel can be selected based on the neighboring relationship between the pixels in the image to be processed. Two possible ways of selecting the reference pixels of the first pixel are given below in conjunction with FIGS. 4 and 5.
  • pixels 1, 3, 6, 7 are the first type of pixels of the image to be processed
  • pixels 2, 4, 5 are the second type of pixels of the image to be processed.
  • the first pixel is the pixel 2 in FIG. 4
  • the distance between the first type of pixels of the image to be processed and the pixel 2 may be selected, and the nearest 4 pixels from the pixel 2 of the image to be processed are selected.
  • the reference pixels of the pixel 2 are the pixels 1, 3, 6, and 7.
  • the reference pixels of the pixel 2 can also be pixels 1, 3, 6, and 7.
  • the reference pixels of the pixel 5 can also be pixels 1, 3, 6, and 7.
  • FIG. 5 is an implementation method of selecting a reference pixel of a first pixel by using an adjacent relationship between pixels as a standard.
  • the pixels 1, 3, 6, and 7 are the first type of pixels of the to-be-processed picture
  • the pixels 2, 4, and 5 are the second type of pixels of the image to be processed.
  • the pixel adjacent to the pixel 2 may be selected from the first type of pixels of the image to be processed based on the adjacent relationship between the first type of pixel and the pixel 2 of the image to be processed.
  • the reference pixels of the pixel 2 are the pixels 1, 3.
  • the pixel adjacent to the pixel 4 may be selected from the first type of pixels of the image to be processed as the pixel 4 based on the adjacent relationship between the first type of pixel and the pixel 4 of the image to be processed.
  • Reference pixel As shown in FIG. 5, the reference pixels of the pixel 4 are the pixels 1, 6.
  • the pixel adjacent to the pixel 5 can be selected from the first type of pixels of the image to be processed based on the adjacent relationship between the first type of pixel and the pixel 5 of the image to be processed.
  • the reference pixels of the pixel 5 are the pixels 1, 3, 6, and 7.
  • the embodiment of the present invention does not specifically limit the number of reference pixels of the first pixel, and may be comprehensively considered in consideration of the accuracy and efficiency of the spot detection algorithm.
  • the upper limit of the number of reference pixels of the first pixel may be preset to be k. If the number of pixels m of the first type of pixels of the image to be processed that can be the reference pixels of the first pixel is greater than k, k pixels may be selected from the m pixels as reference pixels according to a certain preset rule. For example, k pixels may be randomly selected from m pixels as reference pixels of the first pixel.
  • step 320 in a case where the reference pixels of the first pixel and the first pixel satisfy the first preset condition, the first pixel is determined as the spot pixel of the image to be processed.
  • the embodiment of the present invention fully utilizes the pixel relationship between the thumbnail image and the image to be processed to perform spot detection on the image to be processed, which simplifies the spot detection process of the image to be processed.
  • the speckle detection process of the image to be processed described in FIG. 3 does not require random non-directional random access of the entire image of the image to be processed, and is simple to implement.
  • the determination processes of the pixels of the image to be processed are independent of each other, the pixels of the image to be processed are continuously accessed, and the pixels of the image to be processed are processed in parallel.
  • the method of FIG. 2 and the method of FIG. 3 may be implemented independently or in conjunction with one another.
  • the method of FIG. 3 may be continued, so that not only the pixels in the target pixel set of the image to be processed are determined to be speckle pixels, but also the remaining pixels of the image to be processed are also It may be determined as a speckle pixel, which reduces the probability that the speckle pixel in the image to be processed is missed.
  • the manner of setting the first preset condition is not specifically limited in the embodiment of the present invention. For example, it can be based Whether the reference pixel of the first pixel contains a pixel in the target pixel set belonging to the image to be processed (the definition of the target pixel set is referred to above). As another example, it may be set based on a relationship between a reference pixel of the first pixel and a pixel value of the first pixel. Alternatively, it may be a combination of the above two setting methods.
  • the setting manner of the first preset condition is different, and the implementation manner of the step 320 is also different. A specific implementation manner of the step 320 is given below in conjunction with FIG. 6.
  • step 320 can include steps 610-620.
  • step 610 in a case where the reference pixel of the first pixel contains a pixel belonging to the target pixel set of the image to be processed, the first pixel is determined as the spot pixel of the image to be processed.
  • the reference pixel of the first pixel contains a pixel belonging to the target pixel set of the image to be processed, it indicates that the first pixel is adjacent to the region where the target pixel set of the pixel to be processed is located.
  • the pixel adjacent to the region where the target pixel set is located is more likely to be a spot pixel, and such a pixel can be directly determined as a spot pixel, and the spot pixel in the image to be processed is reduced. The probability.
  • step 620 in a case where the reference pixel of the first pixel does not include the pixel belonging to the target pixel set of the image to be processed, the difference between the pixel value of the first pixel and the pixel value of the reference pixel of the first pixel is determined. If the difference between the pixel value of the first pixel and the pixel value of the reference pixel of the first pixel satisfies the second preset condition, the first pixel is determined as the spot pixel of the image to be processed.
  • the reference pixel of the first pixel contains a pixel belonging to the target pixel set of the image to be processed, it indicates that the first pixel is adjacent to the region where the target pixel set of the pixel to be processed is located.
  • the reference pixel of the first pixel does not include the pixel belonging to the target pixel set of the image to be processed, it indicates that the first pixel is not adjacent to the region of the target pixel set of the pixel to be processed (at least one pixel is spaced in the middle).
  • one possible way of determining is to directly determine such pixels as non-spotted pixels of the image to be processed, but this way of determining is likely to miss.
  • the possibility that the first pixel is a speckle pixel is further determined by using a difference between the reference pixel of the first pixel and the first pixel, and the difference between the reference pixel of the first pixel and the first pixel satisfies a preset condition.
  • the first pixel is also determined as a speckle pixel to reduce the probability that the speckle pixel in the image to be processed is missed.
  • step 620 is related to the setting manner of the second preset condition, but the setting manner of the second preset condition is not specifically limited in the embodiment of the present invention, and may be set according to actual needs.
  • the second preset condition may be: if the pixel value of the first pixel is the first image The difference between the pixel values of any one of the reference pixels of the prime is greater than a preset threshold, and the first pixel is determined as the spot pixel of the image to be processed. The difference between the pixel value of the first pixel and the pixel value of the reference pixel is large, indicating that there is a sudden change in the pixel value of the first pixel compared to the peripheral pixel of the first pixel. It is highly probable that a pixel having a sudden change in pixel value is a spot pixel, and such a pixel is determined as a speckle pixel, and the probability of missed detection of the speckle pixel can be reduced.
  • the second preset condition may also adopt other definitions.
  • the second preset condition may be: if the difference between the pixel value of the first pixel and the average value of each pixel value in the reference pixel of the first pixel is greater than a preset threshold, determining the first pixel as a spot of the image to be processed Pixel.
  • FIG. 6 may include step 610 without step 620.
  • step 610 is performed, and conversely, no operation is performed.
  • FIG. 6 may also include step 620 without step 610.
  • step 620 is performed, and conversely, no operation is performed.
  • step 130 no longer requires random access to the entire image of the image to be processed as in the conventional image spot detection mode. Therefore, in some embodiments, the image to be processed can be continuously accessed, such as row by row, column by column, or block by block, and the pixels of the image to be processed are read into the memory for processing, thereby simplifying the access process of the image data and improving Processing efficiency of image data. In addition, since the spot detection process of each pixel in the image to be processed is relatively independent, the pixels stored in the memory can be processed in parallel.
  • the step 130 may include: reading the first image block of the image to be processed from the external memory into the memory at one time; selecting the second image block corresponding to the first image block from the thumbnail image, the second image
  • the block may include pixels in the thumbnail image that are downsampled or sampled by the first image block; and based on the second image block, the speckle pixels in the first image block are determined.
  • the first image block may be composed of one or more rows of pixels in the image to be processed, or may be composed of one or more columns of pixels in the image to be processed, or may have an adjacent relationship in any image region in the image to be processed.
  • the embodiment of the present invention is not specifically limited.
  • the size of the first image block can be flexibly adjusted according to the current capacity of the memory.
  • the first image block can include a plurality of pixels that can be processed in parallel.
  • the parallel processing of pixels in the image to be processed can greatly improve the spot detection efficiency of the image to be processed.
  • FIG. 7 is a schematic structural diagram of an image processing system according to an embodiment of the present invention.
  • the image processing system 700 of FIG. 7 includes a memory 710 and a processor 720.
  • Memory 710 can be used to store programs.
  • the processor 720 can be configured to execute a program stored in the memory. When the program stored in the memory 710 is executed, the processor 720 can be used to perform the speckle detection method described in any of the above embodiments.
  • the computer program product includes one or more computer instructions.
  • the computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable device.
  • the computer instructions can be stored in a computer readable storage medium or transferred from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be from a website site, computer, server or data center Transmission to another website site, computer, server or data center via wired (eg coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (eg infrared, wireless, microwave, etc.).
  • the computer readable storage medium can be any available media that can be accessed by a computer or a data storage device such as a server, data center, or the like that includes one or more available media.
  • the usable medium may be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a digital video disc (DVD)), or a semiconductor medium (such as a solid state disk (SSD)).
  • a magnetic medium for example, a floppy disk, a hard disk, a magnetic tape
  • an optical medium such as a digital video disc (DVD)
  • a semiconductor medium such as a solid state disk (SSD)
  • the disclosed systems, devices, and The method can be implemented in other ways.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

Abstract

提供一种斑点检测方法、系统、存储介质及计算机程序产品。该方法包括：获取待处理图像的缩略图像；对所述缩略图像进行斑点检测，得到所述缩略图像的斑点像素；根据所述缩略图像的斑点像素，确定所述待处理图像的斑点像素。本申请提供的技术方案并非直接对待处理图像进行斑点检测，而是先对待处理图像的缩略图像进行斑点检测，再基于缩略图像的斑点检测结果确定待处理图像的斑点像素。该技术方案有可能提高图像斑点检测效率。

Description

斑点检测方法、系统、存储介质及计算机程序产品
版权申明
本专利文件披露的内容包含受版权保护的材料。该版权为版权所有人所有。版权所有人不反对任何人复制专利与商标局的官方记录和档案中所存在的该专利文件或者该专利披露。
技术领域
本申请涉及图像处理领域，并且更为具体地，涉及一种斑点检测方法、系统、存储介质及计算机程序产品。
背景技术
图像中的斑点(或称斑点噪声)会影响图像的质量。因此，为了提升图像质量，需要对图像中的斑点(speckle)进行斑点检测和斑点滤除。
传统的图像斑点检测方式通过遍历待处理图像中的各像素，直接对待处理图像进行斑点检测，但待处理图像的尺寸一般较大，直接对待处理图像进行斑点检测会导致图像斑点检测效率较低。
发明内容
本申请提供一种斑点检测方法、系统、存储介质及计算机程序产品，可能会提高图像斑点检测的效率。
第一方面，提供一种斑点检测方法，包括：获取待处理图像的缩略图像；对所述缩略图像进行斑点检测，得到所述缩略图像的斑点像素；根据所述缩略图像的斑点像素，确定所述待处理图像的斑点像素。
第二方面，提供一种图像处理系统，包括：存储器，用于存储程序；处理器，用于执行所述存储器中存储的程序，当所述程序被执行时，所述处理器用于执行如第一方面所述的方法。
第三方面，提供一种斑点检测装置，包括用于执行第一方面所述的方法的模块。
第四方面，提供一种计算机可读存储介质，包括计算机指令，当所述计算机指令在计算机上运行时，使得所述计算机执行如第一方面所述的方法。
第五方面，提供一种计算机程序产品，当所述计算机程序产品在计算机上运行时，使得所述计算机执行如第一方面所述的方法。
与待处理图像相比，缩略图像的尺寸较小，缩略图像的斑点检测过程的效率相应较高。进一步地，利用缩略图像的斑点检测结果确定待处理图像中的斑点像素，相当于获知了待处理图像的斑点像素的大致区域的前提下再对待处理图像进行斑点检测，整体上可能会提升待处理图像的斑点检测效率。
附图说明
图1是本发明实施例提供的斑点检测方法的示意性流程图。
图2是图1中的步骤130的一种实现方式的示意性流程图。
图3是图1中的步骤130的另一实现方式的示意性流程图。
图4是以像素之间的距离关系为标准选取第一像素的参考像素的实现方式的示例图。
图5是以像素之间的相邻关系为标准选取第一像素的参考像素的实现方式的示例图。
图6是图3中的步骤320的一种实现方式的示意性流程图。
图7是本发明实施例提供的图像处理系统的示意性结构图。
具体实施方式
本申请可应用于各种需要对图像进行斑点检测和/或斑点滤除的图像处理系统。该图像处理系统例如可以是具有视觉感知功能的无人机、机器人、无人驾驶汽车、潜水器等。
本发明实施例提供一种斑点检测方法，可以提高图像的斑点检测效率。下面结合图1对本发明实施例进行详细描述。
图1是本发明实施例提供的斑点检测方法的示意性流程图。图1的方法可以由上文描述的图像处理系统执行。例如可以由图像处理系统中的处理单元执行。该处理单元例如可以是以下处理单元中的至少一种：中央处理单元(central processing unit，CPU)，数字信号处理器(digital signal processor，DSP)或硬化的IP核(intellectual property core)。图1的方法可以包括步骤110-130，下面分别对步骤110-130进行详细描述。
在步骤110中，获取待处理图像的缩略图像。待处理图像的缩略图像也 可称为待处理图像的小图。缩略图像和待处理图像描绘的对象为相同对象。例如，缩略图像和待处理图像可以描述同一场景或同一人物。缩略图像和待处理图像的不同之处在于缩略图像的分辨率小于待处理图像的分辨率。例如，缩略图像的分辨率为256×256，待处理图像的分辨率为1024×1024。
本发明实施例对缩略图像的获取方式不做具体限定。作为一个示例，可以对待处理图像进行降采样，得到缩略图像。作为另一示例，可以对待处理图像进行抽样，得到缩略图像。换句话说，缩略图像可以是待处理图像的降采样图像或抽样图像。抽样图像的像素可以由待处理图像的部分像素(即待处理图像中的作为抽样点的像素)构成。
在步骤120中，对缩略图像进行斑点检测，得到缩略图像的斑点像素。本发明实施例对缩略图像的斑点检测方式不做具体限定，可以采用任意斑点检测算法对缩略图像进行斑点检测。例如，在一种基于连通域的斑点检测算法中，可以先检测缩略图像中的各像素的像素值。如果缩略图像中的具有相连关系的像素的像素值接近(如像素值之差小于预设差值)，则可以将这些具有相邻关系的像素所在的区域称为连通域。其次，可以判断缩略图像中的各连通域的面积。如果连通域的面积小于预设阈值，则将该连通域中的像素均视为斑点像素。
在一些实施例中，缩略图像的斑点像素可以指缩略图像中的斑点区域中的像素。或者，缩略图像的斑点像素可以指缩略图像中的被判定为斑点或斑点噪声的像素。
在步骤130中，根据缩略图像的斑点像素，确定待处理图像的斑点像素。应理解，步骤130的实现方式有多种，本发明实施例对此不做具体限定。例如，可以将待处理图像中的与缩略图像中的斑点像素对应的像素直接确定为待处理图像的斑点像素。又如，可以根据待处理图像中的与缩略图像的斑点像素对应的像素及其周围像素的像素值，确定待处理图像的斑点像素。后文会结合具体的实施例进行详细描述。
与待处理图像相比，缩略图像的尺寸较小，缩略图像的斑点检测过程的效率相应较高。进一步地，利用缩略图像的斑点检测结果确定待处理图像中的斑点像素，相当于获知了待处理图像的斑点像素的大致区域的前提下再对待处理图像进行斑点检测，整体上可能会提升待处理图像的斑点检测效率。
在一种现有的基于连通域的图像斑点检测方式中，需要对待处理图像的 整个图像的数据进行随机访问，这是由于图像的连通域的形状具有不确定性，可以朝图像的任意方向扩展而引起的。图像处理系统可以对内存中的图像数据进行高效的随机访问。但图像处理系统(尤其是片上图像处理系统)的内存容量一般较小，无法一次性容纳整幅图像的数据，导致图像处理系统需要将整幅图像存储到外部存储器中，并对外部存储器中的图像数据进行随机访问。图像处理系统对外部存储器的随机访问的效率较低，导致传统的图像斑点检测方式的效率较低。
可选地，作为一个实施例，可以将缩略图像的图像数据的大小设定为小于内存的容量(该内存的容量可以指内存的总容量，也可以指内存当前的空闲空间的容量)。在步骤120之前，图1的方法还可包括将缩略图像的图像数据从外部存储器读取到内存中(一次性读取到内存中)；步骤120可包括：在内存中对缩略图像进行斑点检测。假设缩略图像的抽样(或下采样)的比例为1/M，对缩略图像进行斑点检测所需的内存空间为直接对待处理图像进行斑点检测所需的内存空间的1/(M*M)，因此，与待处理图像相比，缩略图像对内存容量的需求较低。由于缩略图像较小，可以一次性读取到内存中，图像处理系统无需随机访问外部存储设备，这将极大程度提升图像的斑点检测效率。
可选地，作为一个实施例，可以将缩略图像的图像数据的大小设定为小于内存的容量(该内存的容量可以指内存的总容量，也可以指内存当前的空闲空间的容量)。步骤110可包括将缩略图像的图像数据从外部存储器一次性读取到内存中。假设缩略图像的抽样(或下采样)的比例为1/M，对缩略图像进行斑点检测所需的内存空间为直接对待处理图像进行斑点检测所需的内存空间的1/(M*M)。进一步地，本发明实施例中，由于缩略图像较小，可以一次性读取到内存中，图像处理系统无需随机访问外部存储设备，这将极大程度提升图像的斑点检测效率。
下面结合图2-图6，对步骤130的具体实现方式进行详细的举例说明。
图2是图1中的步骤130的一种实现方式的示意性流程图。在图2对应的实现方式中，缩略图像的像素为待处理图像的部分像素，即缩略图像的像素可以由待处理图像的部分像素构成。例如，缩略图像可以为待处理图像的抽样图像。图2的方法可以包括步骤210-220，下面分别对步骤210-220进行详细描述。
在步骤210中，根据缩略图像的斑点像素，从待处理图像中选取目标像素集合。
在步骤220中，将目标像素集合中的像素判定为待处理图像的斑点像素。
目标像素集合中的像素可以为与缩略图像的斑点像素具有对应关系的像素。该对应关系可以指：将缩略图像放大至与待处理图像的尺寸相等之后，目标像素集合中的像素在待处理图像中的相对位置与斑点像素在缩略图像中的相对位置相同。在一些实施例中，缩略图像由待处理图像中的部分像素组成，目标像素集合中的像素可以为与缩略图像的斑点像素相同的像素。本发明实施例是将待处理图像中的与缩略图像的斑点像素相同的像素直接判定为待处理图像的斑点像素。
举例说明，假设待处理图像包含像素{p0,p1,p2,p3,p4,p5,p6,p7,p8}，缩略图像包含像素{p0,p3,p6}，且假设经过步骤210，缩略图像中的像素{p3,p6}被判定为斑点像素，则待处理图像的像素{p3,p6}构成了该待处理图像的目标像素集合，且在图2对应的实现方式中，待处理图像的像素{p3,p6}会被直接判定为待处理图像的斑点像素。
缩略图像中的像素由待处理图像的部分像素构成，如果缩略图像中的某个像素被判定为斑点像素，则待处理图像中的与该像素相同的像素很大概率也是斑点像素。图2对应的实现方式充分利用缩略图像和待处理图像的像素关系，将待处理图像中的与缩略图像中的斑点像素相同的像素直接判定为待处理图像的斑点像素，简化了待处理图像的斑点检测方式。
图3是图1中的步骤130的另一实现方式的示意性流程图。在图3对应的实现方式中，缩略图像的像素为待处理图像的部分像素，即缩略图像的像素可以由待处理图像的部分像素构成。例如，缩略图像可以为待处理图像的抽样图像。图3的方法可以包括步骤310-320，下面分别对步骤310-320进行详细描述。
应理解，图3中的第一像素可以为待处理图像中的除目标像素集合(目标像素集合的定义与图2对目标像素集合的定义相同，具体可以参见图2的相关描述)之外的剩余像素中的任意一个像素。为了便于描述，下文将待处理图像中的除目标像素集合之外的剩余像素所形成的像素集合称为待处理图像的第二像素集合，则第一像素为待处理图像的第二像素集合中的任意一个像素。
可选地，在一些实施例中，可以将待处理图像的第二像素集合中的每个像素作为第一像素执行一次如图3所示的流程。可选地，在另一些实施例中，可以仅将待处理图像的第二像素集合中的部分像素作为第一像素执行如图3所示的流程。例如，在某些应用场合，用户并不关心待处理图像的边缘区域是否出现斑点像素。因此，在这种应用场合下，可以无需将待处理图像的第二像素集合中的位于待处理图像的边缘区域的像素作为第一像素执行如图3所示的流程。
在步骤310中，从待处理图像的部分像素中选取第一像素的参考像素。参考像素可以为上述待处理图像的部分像素中的与第一像素相邻的像素，或者待处理图像的部分像素中的距离第一像素最近的一个或多个像素。
需要说明的是，步骤310中的待处理图像的部分像素是指待处理图像中的用于形成缩略图像的像素。为了便于描述，后文将待处理图像的部分像素称为待处理图像的第一类像素，并将待处理图像中的除第一类像素之外的剩余像素称为待处理图像的第二类像素。
第一像素的参考像素的选取方式有多种。例如，可以基于待处理图像中的像素之间的距离关系选取第一像素的参考像素。又如，可以基于待处理图像中的像素之间的相邻关系选取第一像素的参考像素。下面结合图4和图5给出第一像素的参考像素的两种可能的选取方式。
图4是以像素之间的距离关系为标准选取第一像素的参考像素的实现方式的一个示例。参见图4，像素1,3,6,7为待处理图像的第一类像素，像素2,4,5为待处理图像的第二类像素。假设第一像素为图4中的像素2，则可以基于待处理图像的第一类像素与像素2的距离关系，从待处理图像的第一类像素中选取距离像素2最近的4个像素，作为像素2的参考像素。从图4可以看出，像素2的参考像素为像素1,3,6,7。同理，假设第一像素为像素4，则可以基于待处理图像的第一类像素与像素4的距离关系，从待处理图像的第一类像素中选取距离像素4最近的4个像素，作为像素4的参考像素。从图4可以看出，像素2的参考像素也可以为像素1,3,6,7。同理，假设第一像素为像素5，则可以基于待处理图像的第一类像素与像素5的距离关系，从待处理图像的第一类像素中选取距离像素5最近的4个像素，作为像素5的参考像素。从图4可以看出，像素2的参考像素也可以为像素1,3,6,7。
图5是以像素之间的相邻关系为标准选取第一像素的参考像素的实现方 式的一个示例。参见图5所，像素1,3,6,7为待处理图的第一类像素，像素2,4,5为待处理图像的第二类像素。假设第一像素为图4中的像素2，则可以基于待处理图像的第一类像素与像素2的相邻关系，从待处理图像的第一类像素中选取与像素2相邻的像素，作为像素2的参考像素。如图5所示，像素2的参考像素为像素1,3。假设第一像素为像素4，则可以基于待处理图像的第一类像素与像素4的相邻关系，从待处理图像的第一类像素中选取与像素4相邻的像素，作为像素4的参考像素。如图5所示，像素4的参考像素为像素1,6。同理，假设第一像素为像素5，则可以基于待处理图像的第一类像素与像素5的相邻关系，从待处理图像的第一类像素中选取与像素5相邻的像素，作为像素5的参考像素。如图5所示，像素5的参考像素为像素1,3,6,7。
本发明实施例对第一像素的参考像素的个数不做具体限定，可以综合考虑斑点检测算法的精度和效率的需求而定。在一些实施例中，为了简化算法，可以预先设定第一像素的参考像素的个数的上限为k。如果待处理图像的第一类像素中的能够作为第一像素的参考像素的像素个数m大于k，则可以按照一定的预设规则从该m个像素中选取k个像素作为参考像素。例如，可以随机从m个像素中选取k个像素作为第一像素的参考像素。
在步骤320中，在第一像素和第一像素的参考像素满足第一预设条件的情况下，将第一像素判定为待处理图像的斑点像素。
本发明实施例充分利用缩略图像和待处理图像的像素关系对待处理图像进行斑点检测，简化了待处理图像的斑点检测过程。进一步地，图3描述的待处理图像的斑点检测过程无需对待处理图像的整幅图像进行不定向的随机访问，实现简单。此外，图2和图3对应的实现方式中，待处理图像的各像素的判定过程相互独立，可以对待处理图像的像素进行连续访问，并且可以对待处理图像的像素进行并行处理。
应当理解的是，图2的方法和图3的方法可以独立实施，也可以相互接合。作为一个示例，在执行完图2的方法之后，可以继续执行图3的方法，这样一来，不但待处理图像的目标像素集合中的像素会被判定为斑点像素，待处理图像的剩余像素也有可能会被判定为斑点像素，降低待处理图像中的斑点像素被遗漏的概率。
本发明实施例对第一预设条件的设定方式不做具体限定。例如，可以基 于第一像素的参考像素是否包含属于待处理图像的目标像素集合(目标像素集合的定义参见前文)中的像素而定。又如，可以基于第一像素的参考像素与第一像素的像素值的关系设定。或者，可以是以上两种设定方式的组合。第一预设条件的设定方式不同，步骤320的实现方式也就相应有所不同，下面结合图6，给出步骤320的一种具体实现方式。
如图6所示，步骤320可包括步骤610-620。
在步骤610中，在第一像素的参考像素包含属于待处理图像的目标像素集合的像素的情况下，将第一像素判定为待处理图像的斑点像素。
如果第一像素的参考像素包含属于待处理图像的目标像素集合的像素，则表明第一像素与待处理像素的目标像素集合所在区域相邻。与待处理图像的其他像素相比，与目标像素集合所在区域相邻的像素为斑点像素的可能性更大，可以将这类像素直接判定为斑点像素，降低待处理图像中的斑点像素被遗漏的概率。
在步骤620中，在第一像素的参考像素不包含属于待处理图像的目标像素集合的像素的情况下，确定第一像素的像素值与第一像素的参考像素的像素值的差异。如果述第一像素的像素值与第一像素的参考像素的像素值的差异满足第二预设条件，将第一像素判定为待处理图像的斑点像素。
上文指出，如果第一像素的参考像素包含属于待处理图像的目标像素集合的像素，表明第一像素与待处理像素的目标像素集合所在区域相邻。同理，如果第一像素的参考像素不包含属于待处理图像的目标像素集合的像素，表明第一像素与待处理像素的目标像素集合所在区域不相邻(中间至少间隔了一个像素)。对于此类像素，一种可能的判定方式是直接将此类像素判定为待处理图像的非斑点像素，但是，这种判定方式存在漏判的可能。本发明实施例是利用第一像素与第一像素的参考像素的差异进一步判断第一像素为斑点像素的可能性，并在第一像素和第一像素的参考像素的差异满足预设条件的情况下，也将该第一像素判定为斑点像素，以降低待处理图像中的斑点像素被漏判的概率。
应理解，步骤620得出的判定结果与第二预设条件的设定方式有关，但本发明实施例对第二预设条件的设定方式不做具体限定，可以根据实际需要设定。
作为一个示例，第二预设条件可以是：如果第一像素的像素值与第一像 素的参考像素中的任意一个像素的像素值之差大于预设阈值，将第一像素判定为待处理图像的斑点像素。第一像素的像素值与参考像素的像素值的差异较大，说明与第一像素的周边像素相比，第一像素的像素值存在突变。像素值存在突变的像素是斑点像素的可能性较大，将这种像素判定为斑点像素，可以降低斑点像素的漏判的概率。
当然，除了以上定义方式之外，第二预设条件还可以采用其他定义方式。例如，第二预设条件可以是：如果第一像素的像素值与第一像素的参考像素中的各像素值的均值之差均大于预设阈值，将第一像素判定为待处理图像的斑点像素。
上文是以图6包含步骤610和步骤620为例进行说明的。实际上，在一些实施例中，图6可以包含步骤610，而不包含步骤620。例如，当判断第一像素的参考像素包含属于待处理图像的目标像素集合的像素的情况下，执行步骤610，反之，不执行任何操作。可选地，在另一些实施例中，图6也可以包含步骤620，而不包含步骤610。例如，当判断第一像素的参考像素不包含属于待处理图像的目标像素集合的像素的情况下，执行步骤620，反之，不执行任何操作。
从上文描述的实施例可以看出，步骤130的执行过程不再需要像传统图像斑点检测方式那样对待处理图像的整幅图像进行随机访问。因此，在一些实施例中，可以对待处理图像进行连续访问，如逐行、逐列或逐块地将待处理图像的像素读取到内存中进行处理，这样可以简化图像数据的访问过程，提升图像数据的处理效率。此外，由于待处理图像中的各像素的斑点检测过程相对独立，可以对内存中存储的像素进行并行处理。
作为一个示例，步骤130可包括：将待处理图像的第一图像块从外部存储器一次性读取到内存中；从缩略图像中选取与第一图像块对应的第二图像块，第二图像块可以包含缩略图像中的由第一图像块降采样或抽样得到的像素；根据第二图像块，确定第一图像块中的斑点像素。
第一图像块可以由待处理图像中的一行或多行像素构成，也可以由待处理图像中的一列或多列像素构成，也可以由待处理图像中的任意图像区域中的具有相邻关系的多个像素构成，本发明实施例对此不做具体限定。
进一步地，在一些实施例中，第一图像块的尺寸可以根据内存的当前容量灵活调整。
此外，在一些实施例中，第一图像块可以包含多个像素，该多个像素可以并行处理。待处理图像中的像素的并行处理可以极大提高待处理图像的斑点检测效率。
上文中结合图1至图6，详细描述了本发明实施例提供的斑点检测方法，下面将结合图7，详细描述本发明实施例提供的图像处理系统。
图7是本发明实施例提供的图像处理系统的示意性结构图。图7的图像处理系统700包括存储器710和处理器720。存储器710可用于存储程序。处理器720可用于执行所述存储器中存储的程序。当存储器710中存储的程序被执行时，所述处理器720可用于执行上文任一实施例描述的斑点检测方法。
在上述实施例中，可以全部或部分地通过软件、硬件、固件或者其他任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行所述计算机程序指令时，全部或部分地产生按照本发明实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线(digital subscriber line，DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质(例如，软盘、硬盘、磁带)、光介质(例如数字视频光盘(digital video disc，DVD))、或者半导体介质(例如固态硬盘(solid state disk，SSD))等。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和 方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (14)

  1. 一种斑点检测方法，其特征在于，包括：
    获取待处理图像的缩略图像；
    对所述缩略图像进行斑点检测，得到所述缩略图像的斑点像素；
    根据所述缩略图像的斑点像素，确定所述待处理图像的斑点像素。
  2. 如权利要求1所述的方法，其特征在于，所述根据所述缩略图像的斑点像素，确定所述待处理图像的斑点像素，包括：
    根据所述缩略图像的斑点像素，从所述待处理图像中选取目标像素集合，所述目标像素集合中的像素为与所述缩略图像的斑点像素具有对应关系的像素；
    将所述目标像素集合中的像素判定为所述待处理图像的斑点像素。
  3. 如权利要求1所述的方法，其特征在于，所述根据所述缩略图像的斑点像素，确定所述待处理图像的斑点像素，包括：
    从所述部分像素中选取第一像素的参考像素，所述第一像素为所述待处理图像中的除目标像素集合之外的剩余像素中的任意一个像素，所述目标像素集合中的像素为与所述缩略图像的斑点像素具有对应关系的像素，所述参考像素为所述部分像素中的距离所述第一像素最近的一个或多个像素；
    在所述第一像素和所述第一像素的参考像素满足第一预设条件的情况下，将所述第一像素判定为所述待处理图像的斑点像素。
  4. 如权利要求3所述的方法，其特征在于，所述在所述第一像素和所述第一像素的参考像素满足第一预设条件的情况下，将所述第一像素判定为所述待处理图像的斑点像素，包括：
    在所述第一像素的参考像素包含属于所述目标像素集合的像素的情况下，将所述第一像素判定为所述待处理图像的斑点像素；
    和/或，
    在所述第一像素的参考像素不包含属于所述目标像素集合的像素的情况下，确定所述第一像素的像素值与所述第一像素的参考像素的像素值的差异；如果所述述第一像素的像素值与所述第一像素的参考像素的像素值的差异满足第二预设条件，将所述第一像素判定为所述待处理图像的斑点像素。
  5. 如权利要求4所述的方法，其特征在于，所述如果所述第一像素的像素值与所述第一像素的参考像素的像素值的差异满足第二预设条件，将所 述第一像素判定为所述待处理图像的斑点像素，包括：
    如果所述第一像素的像素值与所述第一像素的参考像素中的任意一个像素的像素值之差大于预设阈值，将所述第一像素判定为所述待处理图像的斑点像素。
  6. 如权利要求3-5中任一项所述的方法，其特征在于，所述第一像素的参考像素为所述待处理图像中的与所述第一像素相邻的像素。
  7. 如权利要求1所述的方法，其特征在于，所述根据所述缩略图像的斑点像素，确定所述待处理图像的斑点像素，包括：
    从所述待处理图像中选取目标像素集合中的至少一个像素中的每个像素的相邻像素，所述目标像素集合中的像素为与所述缩略图像的斑点像素具有对应关系的像素；
    将所述每个像素的相邻像素判定为所述待处理图像的斑点像素。
  8. 如权利要求2、3或7所述的方法，其特征在于，所述缩略图像的像素为所述待处理图像的部分像素，所述目标像素集合中的像素为与所述缩略图像的斑点像素相同的像素。
  9. 如权利要求1-5中任一项所述的方法，其特征在于，所述缩略图像的图像数据的大小小于内存的容量，
    在所述对所述缩略图像进行斑点检测之前，所述方法还包括：
    将所述缩略图像读取到所述内存中；
    所述对所述缩略图像进行斑点检测，包括：
    在所述内存中对所述缩略图像进行斑点检测。
  10. 如权利要求1-5中任一项所述的方法，其特征在于，所述根据所述缩略图像的斑点像素，确定所述待处理图像的斑点像素，包括：
    将所述待处理图像的第一图像块从外部存储器一次性读取到内存中；
    从所述缩略图像中选取与所述第一图像块对应的第二图像块，所述第二图像块包含所述缩略图像中的由所述第一图像块降采样或抽样得到的像素；
    根据所述第二图像块，在所述内存中确定所述第一图像块中的斑点像素。
  11. 如权利要求1-5中任一项所述的方法，其特征在于，所述缩略图像是所述待处理图像的降采样图像或抽样图像。
  12. 一种图像处理系统，其特征在于，包括：
    存储器，用于存储程序；
    处理器，用于执行所述存储器中存储的程序，当所述程序被执行时，所述处理器用于执行如权利要求1-11中任一项所述的方法。
  13. 一种计算机可读存储介质，其特征在于，包括计算机指令，当所述计算机指令在计算机上运行时，使得所述计算机执行如权利要求1-11中任一项所述的方法。
  14. 一种计算机程序产品，其特征在于，当所述计算机程序产品在计算机上运行时，使得所述计算机执行如权利要求1-11中任一项所述的方法。
PCT/CN2017/099978 2017-08-31 2017-08-31 斑点检测方法、系统、存储介质及计算机程序产品 WO2019041245A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CN201780004458.6A CN108496202A (zh) 2017-08-31 2017-08-31 斑点检测方法、系统、存储介质及计算机程序产品
PCT/CN2017/099978 WO2019041245A1 (zh) 2017-08-31 2017-08-31 斑点检测方法、系统、存储介质及计算机程序产品
US16/803,022 US20200202510A1 (en) 2017-08-31 2020-02-27 Speckle detection method, system, storage medium, and computer program product

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/099978 WO2019041245A1 (zh) 2017-08-31 2017-08-31 斑点检测方法、系统、存储介质及计算机程序产品

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/803,022 Continuation US20200202510A1 (en) 2017-08-31 2020-02-27 Speckle detection method, system, storage medium, and computer program product

Publications (1)

Publication Number Publication Date
WO2019041245A1 true WO2019041245A1 (zh) 2019-03-07

Family

ID=63344794

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/099978 WO2019041245A1 (zh) 2017-08-31 2017-08-31 斑点检测方法、系统、存储介质及计算机程序产品

Country Status (3)

Country Link
US (1) US20200202510A1 (zh)
CN (1) CN108496202A (zh)
WO (1) WO2019041245A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022174397A1 (zh) * 2021-02-19 2022-08-25 深圳市大疆创新科技有限公司 图像检测方法、装置及存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004294202A (ja) * 2003-03-26 2004-10-21 Seiko Epson Corp 画面の欠陥検出方法及び装置
CN102467741A (zh) * 2010-11-18 2012-05-23 索尼公司 对图像中的斑状区域进行检测的方法和装置
CN103235949A (zh) * 2013-04-12 2013-08-07 北京大学 图像兴趣点检测方法和装置
US20130278761A1 (en) * 2012-04-23 2013-10-24 Xerox Corporation Real-time video triggering for traffic surveilance and photo enforcement applications using near infrared video acquisition
CN103488992A (zh) * 2013-08-28 2014-01-01 北京理工大学 一种面向复杂sar图像场景的溢油检测方法
CN103971357A (zh) * 2013-02-04 2014-08-06 阿里巴巴集团控股有限公司 一种图像中斑点区域提取方法和装置

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006072642A (ja) * 2004-09-01 2006-03-16 Noritsu Koki Co Ltd 観光情報案内装置
JP5541205B2 (ja) * 2011-03-22 2014-07-09 株式会社ニコン 画像処理装置、撮像装置、画像処理プログラムおよび画像処理方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004294202A (ja) * 2003-03-26 2004-10-21 Seiko Epson Corp 画面の欠陥検出方法及び装置
CN102467741A (zh) * 2010-11-18 2012-05-23 索尼公司 对图像中的斑状区域进行检测的方法和装置
US20130278761A1 (en) * 2012-04-23 2013-10-24 Xerox Corporation Real-time video triggering for traffic surveilance and photo enforcement applications using near infrared video acquisition
CN103971357A (zh) * 2013-02-04 2014-08-06 阿里巴巴集团控股有限公司 一种图像中斑点区域提取方法和装置
CN103235949A (zh) * 2013-04-12 2013-08-07 北京大学 图像兴趣点检测方法和装置
CN103488992A (zh) * 2013-08-28 2014-01-01 北京理工大学 一种面向复杂sar图像场景的溢油检测方法

Also Published As

Publication number Publication date
CN108496202A (zh) 2018-09-04
US20200202510A1 (en) 2020-06-25

Similar Documents

Publication Publication Date Title
WO2014055203A1 (en) Video scene detection
US20170070664A1 (en) Method for processing image and electronic apparatus therefor
US9075827B2 (en) Image retrieval apparatus, image retrieval method, and storage medium
JP2013500536A (ja) 立体画像対における遠近感および視差の調整
JP6465215B2 (ja) 画像処理プログラムおよび画像処理装置
US9967473B2 (en) Information processing apparatus, information processing method, and program
US11765469B2 (en) Image capturing apparatus, device, control method, and computer-readable storage medium
JP2015154465A (ja) 表示制御装置、表示制御方法及びプログラム
US11438511B2 (en) Method for controlling electronic device and electronic device thereof
US11170520B2 (en) Image processing apparatus for analyzing an image to detect an object within the image
JP2011248463A (ja) 画像処理装置、画像処理方法、プログラム、及び集積回路
JP4617883B2 (ja) 画像処理装置および方法、プログラム並びに記録媒体
WO2019041245A1 (zh) 斑点检测方法、系统、存储介质及计算机程序产品
US20170300271A1 (en) Memory control apparatus and memory control method
US10922791B2 (en) Image processing apparatus and method
US11755758B1 (en) System and method for evaluating data files
KR101943097B1 (ko) 가상 특징점을 이용한 이미지 정합 방법 및 그 시스템
US9826163B2 (en) Image processing apparatus, control method, and recording medium
JP5616743B2 (ja) 撮像装置および画像処理方法
JP5702269B2 (ja) 並列映像解析装置、並列映像解析方法及び並列映像解析プログラム
JP5762014B2 (ja) 再生装置及び再生装置の制御方法
US20140267432A1 (en) Method to select appropriate window size for local image processing
US9818444B2 (en) Information processing apparatus, control method thereof, and program
US9384527B2 (en) Electronic device and image displaying method
US20180108112A1 (en) Image processing method and apparatus

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: 17923688

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17923688

Country of ref document: EP

Kind code of ref document: A1