CN108665459B - Image blur detection method, computing device and readable storage medium - Google Patents
Image blur detection method, computing device and readable storage medium Download PDFInfo
- Publication number
- CN108665459B CN108665459B CN201810497616.2A CN201810497616A CN108665459B CN 108665459 B CN108665459 B CN 108665459B CN 201810497616 A CN201810497616 A CN 201810497616A CN 108665459 B CN108665459 B CN 108665459B
- Authority
- CN
- China
- Prior art keywords
- contour line
- image
- contour
- spot
- light spot
- 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.)
- Active
Links
Images
Classifications
-
- 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
- G06T7/0012—Biomedical image inspection
- G06T7/0014—Biomedical image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30168—Image quality inspection
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
- G06T2207/30201—Face
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Quality & Reliability (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an image blur detection method, which is suitable for being executed in computing equipment and comprises the following steps: receiving an image to be detected; intercepting a light spot image from the image, wherein the light spot image comprises at least one light spot; extracting the contour line of each light spot in the light spot image; for each extracted contour line, calculating the Hausdorff distance from the contour line to the corresponding standard contour line; judging whether the calculated Hausdorff distance is smaller than a distance threshold of the standard contour line; and if the proportion of the contour lines with the Hausdorff distance from the corresponding standard contour line to the distance threshold of the standard contour line in all the extracted contour lines exceeds a preset proportion, determining that the image has blur, and determining that the blur type is motion blur. The invention also discloses a corresponding computing device and a readable storage medium.
Description
Technical Field
The present invention relates to the field of image processing technologies, and in particular, to an image blur detection method, a computing device, and a readable storage medium.
Background
With the increasing concern of people on internet security, the application of using biological features such as human faces and irises to identify the identity of users is increasing. The mode of shooting the face/periocular/iris image for identification has the advantages of high safety and high accuracy. Wherein, whether the face/periocular/iris image is clear is the key.
Generally, the image is disturbed by external factors such as shake, motion and focus during the photographing process to generate blur. If the fuzzy face/eye/iris image is identified, the identification error of the user identity is easily caused, thereby bringing trouble to the user. Therefore, it is necessary to perform blur detection on an image for identification such as a face/periocular/iris image.
Existing image motion blur detection methods can be broadly divided into two categories: one is to estimate the blur degree of the whole image, and the other is to divide the image into a plurality of sub-regions and estimate the blur degree of each sub-region so as to detect whether the image is blurred.
Therefore, a more advanced image blur detection scheme is urgently required.
Disclosure of Invention
To this end, the present invention provides an image blur detection method, a computing device and a readable storage medium in an effort to solve or at least alleviate at least one of the problems identified above.
According to an aspect of the present invention, there is provided an image blur detection method, adapted to be executed in a computing device, the method comprising the steps of: receiving an image to be detected; intercepting a light spot image from the image, wherein the light spot image comprises at least one light spot; extracting the contour line of each light spot in the light spot image; for each extracted contour line, calculating the Hausdorff distance from the contour line to the corresponding standard contour line; judging whether the calculated Hausdorff distance is smaller than a distance threshold of a standard contour line, wherein the distance thresholds of the standard contour line and the standard contour line are obtained in advance based on a plurality of blur-free light spot images; and if the proportion of the contour lines with the Hausdorff distance from the corresponding standard contour line to the distance threshold of the standard contour line in all the extracted contour lines exceeds a preset proportion, determining that the image has blur, and determining that the blur type is motion blur.
Alternatively, in the image blur detection method according to the present invention, the step of intercepting the light spot image from the image includes: intercepting an interested area from the image, wherein the interested area contains a light spot image; acquiring the light spot characteristics of the at least one light spot based on the region of interest; and intercepting the spot image according to the spot characteristics of at least one spot.
Optionally, in the image blur detection method according to the present invention, the step of acquiring the spot characteristics of at least one spot based on the region of interest includes: carrying out threshold segmentation on the region of interest to obtain a binary image of the region of interest; and determining the spot characteristics of at least one spot according to the binary image.
Optionally, in the image blur detection method according to the present invention, the light spot feature of at least one light spot includes a minimum bounding rectangle surrounding the at least one light spot, and the step of intercepting the light spot image according to the light spot feature of the at least one light spot includes: intercepting the spot image according to a minimum bounding rectangle that encompasses the at least one spot.
Optionally, in the image blur detection method according to the present invention, the spot characteristics of at least one spot further include a minimum circumscribed rectangle aspect ratio and an area of each spot, and the method further includes the steps of: after the light spot characteristics of at least one light spot are obtained based on the region of interest, whether the length-width ratio of the minimum circumscribed rectangle of each light spot meets a first light spot screening condition is judged; if the proportion of the light spots, the length-width ratio of which does not meet the first light spot screening condition, in all the light spots exceeds the first light spot proportion, determining that the image is fuzzy, and determining that the fuzzy type is motion fuzzy; if the ratio of the minimum length-width ratio of the circumscribed rectangle to the light spots meeting the first light spot screening condition in all the light spots exceeds the first light spot ratio, judging whether the area of each light spot meets the second light spot screening condition or not; and if the proportion of the light spots with the areas not meeting the second light spot screening condition in all the light spots exceeds the second light spot proportion, determining that the image is fuzzy, and determining that the fuzzy type is focus fuzzy.
Optionally, in the image blur detection method according to the present invention, the method further comprises the steps of: after extracting the contour line of each light spot in the light spot image, acquiring the contour line characteristics of each contour line, wherein the contour line characteristics of each contour line comprise the perimeter and the area of the contour line and the length-width ratio of the minimum circumscribed rectangle; for each contour line, judging whether the perimeter of the contour line meets a first contour line screening condition, whether the area meets a second contour line screening condition and whether the length-width ratio of the minimum circumscribed rectangle meets a third contour line screening condition; and if the proportion of the contour line which does not meet any one of the first contour line screening condition, the second contour line screening condition and the third contour line screening condition in all the extracted contour lines exceeds the contour line proportion, determining that the image has blur, and determining that the type of the blur is focus blur.
Optionally, in the image blur detection method according to the present invention, the method further comprises the steps of: for each extracted contour line, respectively calculating the Hausdorff distance from the contour line to each standard contour line before calculating the Hausdorff distance from the contour line to the corresponding standard contour line; and selecting the standard contour line with the minimum Hausdorff distance as the standard contour line corresponding to the contour line.
Optionally, in the image blur detection method according to the present invention, the method further includes a step of obtaining at least one standard contour line and a distance threshold value of each standard contour line based on the plurality of blur-free spot images, including: extracting the contour line of each light spot in each non-fuzzy light spot image; dividing the extracted contour lines into at least one contour line set; for each contour line set, a standard contour line representing the contour line set is obtained, and the distance threshold of the standard contour line is calculated based on the contour line set.
Alternatively, in the image blur detection method according to the present invention, the step of dividing the extracted plurality of contour lines into at least one contour line set includes: calculating the Hausdorff distance between every two of the extracted contour lines; detecting whether contour lines which are not divided into any contour line set exist in the extracted contour lines; if so, constructing a contour line set in contour lines which are not divided into any contour line set according to the calculated Hausdorff distance; and repeating the steps of detecting the contour lines and constructing the contour line sets until no contour lines which are not divided into any contour line set exist in the plurality of extracted contour lines.
Alternatively, in the image blur detection method according to the present invention, the step of constructing one contour line set among contour lines not divided into any contour line set based on the calculated hausdorff distance includes: selecting two contour lines with the minimum Hausdorff distance between every two contour lines in contour lines which are not added into any contour line set, and forming a contour line set; in the contour lines which are not added into any contour line set, whether a contour line with the Hausdorff distance smaller than a preset threshold exists in any contour line in the newly formed contour line set is searched; if the contour line set exists, respectively calculating the sum of the Hausdorff distances from each searched contour line to all contour lines in the newly formed contour line set, and adding the contour line with the minimum sum of the Hausdorff distances to the newly formed contour line set; and repeating the steps of searching the contour lines, calculating the sum of the Hausdorff distances and adding the contour lines until no contour line with the Hausdorff distance smaller than a preset threshold value to any contour line in the newly formed contour line set exists in the contour lines without any contour line set.
Optionally, in the image blur detection method according to the present invention, the predetermined threshold is 0.15 to 0.19.
Alternatively, in the image blur detection method according to the present invention, the step of acquiring a standard contour line representing the set of contour lines includes: in the contour line set, the step of selecting one alternative standard contour line is repeated until the selected alternative standard contour lines reach a preset number; merging the selected candidate standard contour lines with a predetermined number into one standard contour line; wherein the step of selecting one of the candidate standard contour lines comprises: for each contour line which is not selected as the alternative standard contour line in the contour line set, calculating the sum of the Hausdorff distances from the contour line which is not selected as the alternative standard contour line except the contour line to the contour line and all the alternative standard contour lines; and selecting the contour line with the minimum sum of the Hausdorff distances as an alternative standard contour line.
Optionally, in the image blur detection method according to the present invention, the step of calculating the distance threshold of the standard contour line based on the contour line set includes: respectively calculating the Hausdorff distance from each contour line in the contour line set to the standard contour line; the maximum Housdov distance is selected as the distance threshold for the standard contour line.
Alternatively, in the image blur detection method according to the present invention, the hausdorff distance is calculated by the following formula:
HD(A,B)＝max{h(A,B),h(B,A)}
wherein
A＝{pa1,pa2,pa3,......,pam}B＝{pb1,pb2,pb3,......,pbn}
HD (A, B) is the Hausdorff distance between A and B, h (A, B) and h (A, B) are the unidirectional Hausdorff distance from A to B and the unidirectional Hausdorff distance from B to A, A and B are two sets containing finite pixel points, | pa-pbI is pixel point paAnd pixel point pbEuclidean distance between, | | pb-paI is pixel point pbAnd pixel point paThe euclidean distance between.
Alternatively, in the image blur detection method according to the present invention, the step of extracting the contour line of the light spot includes: the isocline algorithm is used to extract the contour of the spot.
Optionally, in the image blur detection method according to the present invention, the image is a near-infrared image of the eye circumference, or the iris, or the face.
According to another aspect of the present invention, there is provided a computing device comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for performing any of the image blur detection methods according to the present invention.
According to still another aspect of the present invention, there is provided a readable storage medium storing a program, the program including instructions that, when executed by a computing device, cause the computing device to perform any one of the image blur detection methods according to the present invention.
According to the image blur detection scheme, at least one standard contour line is obtained in advance based on a plurality of non-blur light spot images, the contour line of each light spot in the image to be detected is extracted, each contour line is compared with the standard contour line corresponding to the contour line based on the Hausdorff distance, so that the image blur detection is realized, the accuracy is high, the calculated amount is small, and the blur types of the image can be distinguished to be focus blur and motion blur.
Drawings
To the accomplishment of the foregoing and related ends, certain illustrative aspects are described herein in connection with the following description and the annexed drawings, which are indicative of various ways in which the principles disclosed herein may be practiced, and all aspects and equivalents thereof are intended to be within the scope of the claimed subject matter. The above and other objects, features and advantages of the present disclosure will become more apparent from the following detailed description read in conjunction with the accompanying drawings. Throughout this disclosure, like reference numerals generally refer to like parts or elements.
FIG. 1 schematically illustrates a block diagram of a computing device 100; and
FIG. 2 schematically shows a flow diagram of an image blur detection method 200 according to an embodiment of the invention;
fig. 3A and 3B respectively schematically show a region of interest and a binary image thereof according to an embodiment of the present invention;
FIG. 4 schematically shows a schematic view of the contour of a spot of light according to an embodiment of the invention;
FIG. 5 schematically illustrates a flow chart of a method 500 for partitioning a set of contour lines according to an embodiment of the present invention; and
FIG. 6 illustrates a flow diagram of a method 600 of constructing a set of contour lines, according to one embodiment of the invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
Fig. 1 exemplarily illustrates a block diagram of a computing device 100. The computing device 100 may be implemented as a server, such as a file server, a database server, an application server, a web server, and the like, or as a personal computer including desktop and notebook computer configurations. Moreover, computing device 100 may also be implemented as part of a small-form factor portable (or mobile) electronic device, such as a cellular telephone, a Personal Digital Assistant (PDA), a personal media player device, a wireless web-browsing device, a personal headset device, an application-specific device, or a hybrid device that include any of the above functions.
In a basic configuration 102, computing device 100 typically includes system memory 106 and one or more processors 104. A memory bus 108 may be used for communication between the processor 104 and the system memory 106.
Depending on the desired configuration, the processor 104 may be any type of processing, including but not limited to: the processor 104 may include one or more levels of cache, such as a level one cache 110 and a level two cache 112, a processor core 114, and registers 116. the example processor core 114 may include an Arithmetic Logic Unit (ALU), a Floating Point Unit (FPU), a digital signal processing core (DSP core), or any combination thereof.
Depending on the desired configuration, system memory 106 may be any type of memory, including but not limited to: volatile memory (such as RAM), non-volatile memory (such as ROM, flash memory, etc.), or any combination thereof. System memory 106 may include an operating system 120, one or more programs 122, and program data 124. In some implementations, the program 122 can be configured to execute instructions on an operating system by one or more processors 104 using program data 124.
Computing device 100 may also include an interface bus 140 that facilitates communication from various interface devices (e.g., output devices 142, peripheral interfaces 144, and communication devices 146) to the basic configuration 102 via the bus/interface controller 130. The example output device 142 includes a graphics processing unit 148 and an audio processing unit 150. They may be configured to facilitate communication with various external devices, such as a display or speakers, via one or more a/V ports 152. Example peripheral interfaces 144 may include a serial interface controller 154 and a parallel interface controller 156, which may be configured to facilitate communication with external devices such as input devices (e.g., keyboard, mouse, pen, voice input device, touch input device) or other peripherals (e.g., printer, scanner, etc.) via one or more I/O ports 158. An example communication device 146 may include a network controller 160, which may be arranged to facilitate communications with one or more other computing devices 162 over a network communication link via one or more communication ports 164.
A network communication link may be one example of a communication medium. Communication media may typically be embodied by computer readable instructions, data structures, program modules, and may include any information delivery media, such as carrier waves or other transport mechanisms, in a modulated data signal. A "modulated data signal" may be a signal that has one or more of its data set or its changes made in such a manner as to encode information in the signal. By way of non-limiting example, communication media may include wired media such as a wired network or private-wired network, and various wireless media such as acoustic, Radio Frequency (RF), microwave, Infrared (IR), or other wireless media. The term computer readable media as used herein may include both storage media and communication media.
One or more programs 122 of computing device 100 include instructions for performing any of the image blur detection methods according to the present invention. Fig. 2 illustrates a flow diagram of an image blur detection method 200 according to an embodiment of the present invention.
As shown in fig. 2, the image blur detection method 200 starts at step S210. In step S210, an image to be detected is received. The image is typically acquired at an image acquisition device. According to one embodiment of the invention, the computing device 100 may communicate with the image capture device via one or more communication ports 164 over the network communication link described above to obtain from the image capture device a near-infrared image captured by the image capture device that includes an iris, such as a near-infrared image of a human face, a periocular region, or an iris. The image capture device may generally include a near infrared light source, an optical lens, and an image sensor.
Then, in step S220, a spot image is intercepted from the image, and the spot image contains at least one spot. It will be appreciated that these spots are formed by the reflection of ambient bright objects (e.g., near infrared light sources) onto the pupil.
According to one embodiment of the invention, the spot image may be intercepted by:
the region of interest is first cut out from the image. The region of interest contains an image of the spot, which is typically the pupil region, and can be intercepted by locating the pupil in the image.
Spot characteristics of at least one spot may then be acquired based on the region of interest. Specifically, a region of interest is subjected to threshold segmentation to obtain a binary image of the region of interest, so as to separate the light spot from the background. The threshold segmentation method may be a threshold-based segmentation method such as fixed threshold segmentation and adaptive threshold image segmentation, a region-based segmentation method such as a watershed method, or a geometric active contour model-based segmentation method such as a level set method, which is not limited by the present invention.
Fig. 3A and 3B respectively illustrate a schematic diagram of a region of interest and a binary image thereof according to an embodiment of the present invention, in the binary image shown in fig. 3B, a white portion is a light spot, and a black portion is a background.
The spot characteristics of at least one spot can then be determined from the binary image of the region of interest. In general, the spot characteristics of the at least one spot may include a minimum bounding rectangle surrounding the at least one spot, a minimum bounding rectangle aspect ratio and area for each spot, and so on.
According to another embodiment of the present invention, after acquiring the spot characteristics of at least one spot based on the region of interest, it may be determined whether the aspect ratio of the minimum bounding rectangle of each spot satisfies a first spot screening condition, where the aspect ratio of the minimum bounding rectangle is smaller than the ratio of the length to the width of the spot (e.g., 1.5).
If the proportion of the light spots with the minimum circumscribed rectangle length-width ratio which does not meet the first light spot screening condition in all the light spots exceeds a first light spot proportion (generally 1/2), determining that the image has the blur, and determining that the blur type is the motion blur. If the ratio of the light spots with the minimum length-width ratio of the circumscribed rectangle meeting the first light spot screening condition to all the light spots exceeds the first light spot ratio, whether the area of each light spot meets a second light spot screening condition is continuously judged, wherein the second light spot screening condition is that the area is between the minimum light spot area value and the maximum light spot area value (for example, between 40 pixels and 200 pixels), and comprises the minimum light spot area value and the maximum light spot area value.
If the proportion of the light spots with the areas which do not meet the second light spot screening condition in all the light spots exceeds a second light spot proportion (generally 1/2), the image is determined to have the blur, and the blur type is the focus blur.
And if the proportion of the light spots with the areas meeting the second light spot screening condition in all the light spots exceeds the second light spot proportion, intercepting the light spot image according to the light spot characteristics of at least one light spot. Specifically, the light spot image may be intercepted according to a minimum bounding rectangle surrounding at least one light spot, for example, the intercepted light spot image is a rectangular image surrounding at least the minimum bounding rectangle.
After the spot image is captured, in step S230, the contour line of each spot in the spot image is extracted. According to one embodiment of the present invention, the isocline algorithm may be used to extract the contour of the spot. The isocline algorithm is the prior art, and the basic principle is to find points with the same gray value in a gray image to be connected into a contour line, which is not described in detail in the invention.
Fig. 4 is a schematic diagram illustrating the contour lines of the light spots according to an embodiment of the present invention, and the light spot image shown in fig. 4 includes 2 light spots, and 2 contour lines (indicated by black lines) are extracted.
After extracting the contour line of each spot in the spot image, the extracted contour lines may be subjected to a process of averaging and normalizing according to an embodiment of the present invention.
According to another embodiment of the invention, after the contour line of each spot in the spot image is extracted, the contour line feature of each contour line may be further obtained, and the contour line feature of each contour line may include the aspect ratio, the area, the perimeter, and the like of the minimum circumscribed rectangle of the one contour line. Whether the image is blurred or not can be judged according to the area and the circumference of the outline.
According to one embodiment of the present invention, for each contour line, it is determined whether the circumference of the contour line satisfies a first contour line filtering condition, whether the area satisfies a second contour line filtering condition, and whether the minimum circumscribed rectangle aspect ratio satisfies a third contour line filtering condition. The first contour filtering condition may be that the circumference is between a contour circumference minimum and a contour circumference maximum (e.g., between 30 pixels and 150 pixels), including the contour circumference minimum and the contour circumference maximum. The second contour line filtering condition may be that the area is between a contour line area minimum value and a contour line area maximum value (e.g., between 40 pixels and 200 pixels), including the contour line area minimum value and the contour line area maximum value. The third contour screening condition may be that the minimum bounding rectangle aspect ratio is less than the contour length-to-width ratio (e.g., 1.5).
If the proportion of the contour lines not satisfying any of the first contour line filtering condition, the second contour line filtering condition, and the third contour line filtering condition in all the extracted contour lines exceeds the contour line proportion (generally 1/2), it is determined that the image has blur and the type of blur is focus blur.
If the proportion of the contour lines not satisfying any of the first contour line filtering condition, the second contour line filtering condition, and the third contour line filtering condition in all the extracted contour lines does not exceed the contour line proportion, the process may proceed to step S240.
In step S240, for each contour line extracted, the hausdorff distance of the contour line to its corresponding standard contour line may be calculated. According to one embodiment of the invention, the hausdorff distance may be calculated by the following formula:
HD(A,B)＝max{h(A,B),h(B,A)}
wherein
A＝{pa1,pa2,pa3,......,pam}B＝{pb1,pb2,pb3,......,pbn}
HD (A, B) is the Hausdorff distance between A and B (also called bidirectional Hausdorff distance), h (A, B) and h (A, B) are respectively the unidirectional Hausdorff distance from A to B and the unidirectional Hausdorff distance from B to A, A and B are respectively two sets containing finite pixel points, | | pa-pbI is pixel point paAnd pixel point pbThe euclidean distance between them,||pb-pai is pixel point pbAnd pixel point paThe euclidean distance between.
According to an embodiment of the present invention, the image blur detection method 200 may further include the steps of: and obtaining at least one standard contour line and a distance threshold value of each standard contour line in advance based on the plurality of blur-free light spot images. Therefore, for each extracted contour line, before calculating the hounsfield distance from the contour line to the corresponding standard contour line, the standard contour line corresponding to the contour line needs to be determined. Specifically, the hausdorff distance from the contour line to each standard contour line may be calculated, and finally, the standard contour line with the smallest hausdorff distance may be selected as the standard contour line corresponding to the contour line.
After the hausdorff distance from the contour line to the corresponding standard contour line is calculated, it may be determined whether the calculated hausdorff distance is less than the distance threshold of the standard contour line in step S240.
Finally, in step S250, if the proportion of the contour lines whose hausdorff distances to the corresponding standard contour line are not less than the distance threshold of the standard contour line in all the extracted contour lines exceeds a predetermined proportion, it is determined that the image has a blur, and the type of the blur is a motion blur. The predetermined ratio can be adjusted by the user according to the actual situation, and for example, it can be 1 divided by the number of all contour lines extracted, or it can be 1/2.
And if the proportion of the contour lines of which the Housdov distances to the corresponding standard contour lines are not less than the distance threshold of the standard contour lines in all the extracted contour lines is not more than a preset proportion, determining that the image has no motion blur.
The following describes a process of obtaining at least one standard contour line based on a plurality of blur-free spot images and a distance threshold value of each standard contour line in detail.
First, for a plurality of blur-free spot images, the contour line of each spot in each blur-free spot image can be extracted. In particular, the isocline algorithm can be used to extract the contour lines. The non-fuzzy light spot images are artificially screened light spot image samples which are not considered to be fuzzy. The method can be similar to the step of intercepting the light spot image from the image to be detected, and is intercepted from the human-screened near-infrared image which is not considered to have fuzzy human face or eye circumference or iris, and the size of the image is the same as that of the light spot image. The present invention will not be described in detail herein.
Then, the extracted contour lines are divided into at least one contour line set.
FIG. 5 illustrates a flow diagram of a method 500 for contour line set partitioning, according to one embodiment of the present invention. As shown in FIG. 5, the contour line set division method 500 begins at step S510.
In step S510, housdov distances between two of the extracted plurality of contour lines are calculated. Then, in step S520, it is detected whether there is a contour line that is not divided into any contour line set among the extracted plurality of contour lines. If yes, the process proceeds to step S530, and a contour line set is constructed among contour lines that are not divided into any contour line sets according to the calculated hausdorff distance. After the contour line set is constructed, the process returns to step S520 to continue the detection, and if it is detected that there still exists a contour line that is not divided into any contour line set among the extracted contour lines, the process still proceeds to step S530, and a contour line set is constructed among the contour lines that are not divided into any contour line set according to the calculated hausdorff distance. The above steps S520 and S530 are repeated until there is no contour line that is not divided into any contour line set among the plurality of extracted contour lines.
According to one embodiment of the present invention, if it is detected that, among the extracted contour lines, only two contour lines that are not divided into any contour line set exist, the two contour lines are directly divided into one contour line set. According to another embodiment of the present invention, if it is detected that only one contour line that is not divided into any contour line set exists among the extracted contour lines, the contour line is directly discarded, and it is considered that there is no contour line that is not divided into any contour line set.
The process of constructing one contour line set among contour lines that are not classified into any contour line set based on the calculated hausdorff distance will be described in detail below.
FIG. 6 illustrates a flow diagram of a method 600 of constructing a set of contour lines, according to one embodiment of the invention. As shown in FIG. 6, the method 600 of constructing a set of contour lines begins at step S610.
In step S610, two contour lines with the minimum hausdorff distance between two contour lines are selected from the contour lines not added to any contour line set, and a contour line set is newly formed.
Then, in step S620, it is checked whether there is a contour line whose hausdorff distance to any contour line in the newly composed contour line set is smaller than a predetermined threshold value, among contour lines to which no contour line set is added. The predetermined threshold value is typically between 0.15 and 0.19 (including 0.15 and 0.19).
If so, in step S630, the sum of the housdov distances from each of the searched contour lines to all contour lines in the newly composed contour line set is calculated.
Assume that the newly composed contour line set P ═ { l ═1,l2,......,li,......,lnFor each contour line obtained by searching
The strip of contour line
The sum of the Hausdorff distances to all the contours in the newly composed contour set P is:
then, in step S640, the contour line in which the sum of the hausdorff distances is minimum is added to the newly composed contour line set. Of course, if only one contour line is found, the contour line is directly added to the newly formed contour line set.
Repeating the steps S620, S630 and S640 until no contour line with the Housdov distance to any contour line in the newly formed contour line set being less than the predetermined threshold exists in the contour lines without any contour line set, which means that the newly formed contour line set is completely constructed.
After each extracted contour line is divided into corresponding contour line sets, for each contour line set, a standard contour line representing the contour line set can be obtained, and the distance threshold of the standard contour line is calculated based on the contour line set.
Specifically, in the contour line set, the step of selecting one candidate standard contour line may be repeated until the number of candidate standard contour lines selected reaches a predetermined number. The predetermined number is usually 1 to 3.
The step of selecting an alternative standard contour line may be as follows:
and for each contour line which is not selected as the alternative standard contour line in the contour line set, calculating the sum of the Hausdorff distances from the contour line which is not selected as the alternative standard contour line except the contour line to the contour line and all the alternative standard contour lines.
Assuming that the set of contour lines is P, the contour lines selected as the candidate standard contour lines in the set of contour lines P constitute a set Q. For each contour line l not selected as the candidate standard contour line in the contour line set P, the set S ═ P-Q- { l } of contour lines not selected as the candidate standard contour lines other than the contour line l, then the sum of the hausdov distances from the contour line l not selected as the candidate standard contour line other than the contour line l to the contour line l and all candidate standard contour lines is:
the contour line in which the sum of the hausdorff distances is the smallest may be selected as one of the candidate standard contour lines.
And finally, combining the selected candidate standard contour lines with a preset number into a standard contour line representing the contour line set. Specifically, assume that an alternative standard contour l is picked outq1、lq2And lq3It can be understood that each candidate standard contour line is a set containing limited pixel points, and thus a standard contour line obtained by combination is lq1+lq2+lq3That is, the union of the pixel points contained in the three candidate standard contour lines.
According to an embodiment of the present invention, if the number of contour lines included in the contour line set is less than or equal to the predetermined number, all contour lines included in the contour line set are directly merged to obtain a standard contour line representing the contour line set. For example, if the predetermined number is 3 and the contour line set only includes two contour lines, the two contour lines are merged to obtain a standard contour line representing the contour line set.
After a standard contour line representing the contour line set is obtained, the Hausdorff distance from each contour line in the contour line set to the standard contour line can be respectively calculated, and the maximum Hausdorff distance is selected as the distance threshold of the standard contour line.
In summary, according to the image blur detection scheme of the present invention, at least one standard contour line with high referential degree can be obtained in advance based on a plurality of blur-free light spot images, so that a contour line of each light spot in an image to be detected can be extracted, each contour line is compared with the standard contour line corresponding to the contour line based on the hausdorff distance, thereby realizing blur detection on the image, and the image blur detection scheme has high accuracy and small calculation amount, and can distinguish whether the blur type of the image is focus blur or motion blur.
It should be understood that the various techniques described herein may be implemented in connection with hardware or software or, alternatively, with a combination of both. Thus, the methods and apparatus of the present invention, or certain aspects or portions thereof, may take the form of program code (i.e., instructions) embodied in tangible media, such as floppy diskettes, CD-ROMs, hard drives, or any other machine-readable storage medium, wherein, when the program is loaded into and executed by a machine, such as a computer, the machine becomes an apparatus for practicing the invention.
In the case of program code execution on programmable computers, the computing device will generally include a processor, a storage medium readable by the processor (including volatile and non-volatile memory and/or storage elements), at least one input device, and at least one output device. Wherein the memory is configured to store program code; the processor is configured to perform the various methods of the present invention according to instructions in the program code stored in the memory.
The present invention may further comprise: a9, the image blur detection method according to A8, wherein the step of dividing the extracted contour lines into at least one contour line set includes: calculating the Hausdorff distance between every two of the extracted contour lines; detecting whether contour lines which are not divided into any contour line set exist in the extracted contour lines; if so, constructing a contour line set in contour lines which are not divided into any contour line set according to the calculated Hausdorff distance; and repeating the steps of detecting the contour lines and constructing the contour line sets until no contour lines which are not divided into any contour line set exist in the plurality of extracted contour lines. A10, the image blur detection method according to a9, wherein the step of constructing one contour line set among contour lines not classified into any contour line set according to the calculated hausdorff distance includes: selecting two contour lines with the minimum Hausdorff distance between every two contour lines in contour lines which are not added into any contour line set, and forming a contour line set; in the contour lines which are not added into any contour line set, whether a contour line with the Hausdorff distance smaller than a preset threshold exists in any contour line in the newly formed contour line set is searched; if the contour line set exists, respectively calculating the sum of the Hausdorff distances from each searched contour line to all contour lines in the newly formed contour line set, and adding the contour line with the minimum sum of the Hausdorff distances to the newly formed contour line set; and repeating the steps of searching the contour lines, calculating the sum of the Hausdorff distances and adding the contour lines until no contour line with the Hausdorff distance smaller than a preset threshold value to any contour line in the newly formed contour line set exists in the contour lines without any contour line set. A11, the image blur detection method according to A10, wherein the predetermined threshold is 0.15 to 0.19. A12, the image blur detection method according to any one of a9-11, wherein the step of obtaining a standard contour line representing the set of contour lines includes: in the contour line set, the step of selecting one alternative standard contour line is repeated until the selected alternative standard contour lines reach a preset number; merging the selected candidate standard contour lines with a predetermined number into one standard contour line; wherein the step of selecting one of the candidate standard contour lines comprises: for each contour line which is not selected as the alternative standard contour line in the contour line set, calculating the sum of the Hausdorff distances from the contour line which is not selected as the alternative standard contour line except the contour line to the contour line and all the alternative standard contour lines; and selecting the contour line with the minimum sum of the Hausdorff distances as an alternative standard contour line. A13, the image blur detection method according to any one of a9-12, wherein the step of calculating the distance threshold of the standard contour line based on the set of contour lines comprises: respectively calculating the Hausdorff distance from each contour line in the contour line set to the standard contour line; the maximum Housdov distance is selected as the distance threshold for the standard contour line. A14, the image blur detection method of any one of a9-13, wherein the hausdorff distance is calculated by the following formula:
HD(A,B)＝max{h(A,B),h(B,A)}
wherein
A＝{pa1,pa2,pa3,......,pam}B＝{pb1,pb2,pb3,......,pbn}
HD (A, B) is the Hausdorff distance between A and B, h (A, B) and h (A, B) are the unidirectional Hausdorff distance from A to B and the unidirectional Hausdorff distance from B to A, A and B are two sets containing finite pixel points, | pa-pbI is pixel point paAnd pixel point pbEuclidean distance between, | | pb-paI is pixel point pbAnd pixel point paThe euclidean distance between. A15 the image blur detection method of any one of a1-14, wherein the step of extracting the contour line of the spot includes: the isocline algorithm is used to extract the contour of the spot. A16, the image blur detection method according to any one of A1-15, wherein the image is a near infrared image of the eye circumference, or iris, or human face.
By way of example, and not limitation, computer readable media may comprise computer storage media and communication media. Computer-readable media includes both computer storage media and communication media. Computer storage media store information such as computer readable instructions, data structures, program modules or other data. Communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media. Combinations of any of the above are also included within the scope of computer readable media.
It should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be interpreted as reflecting an intention that: that the invention as claimed requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.
Those skilled in the art will appreciate that the modules or units or components of the devices in the examples disclosed herein may be arranged in a device as described in this embodiment or alternatively may be located in one or more devices different from the devices in this example. The modules in the foregoing examples may be combined into one module or may be further divided into multiple sub-modules.
Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
Furthermore, those skilled in the art will appreciate that while some embodiments described herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the following claims, any of the claimed embodiments may be used in any combination.
Furthermore, some of the described embodiments are described herein as a method or combination of method elements that can be performed by a processor of a computer system or by other means of performing the described functions. A processor having the necessary instructions for carrying out the method or method elements thus forms a means for carrying out the method or method elements. Further, the elements of the apparatus embodiments described herein are examples of the following apparatus: the apparatus is used to implement the functions performed by the elements for the purpose of carrying out the invention.
As used herein, unless otherwise specified the use of the ordinal adjectives "first", "second", "third", etc., to describe a common object, merely indicate that different instances of like objects are being referred to, and are not intended to imply that the objects so described must be in a given sequence, either temporally, spatially, in ranking, or in any other manner.
While the invention has been described with respect to a limited number of embodiments, those skilled in the art, having benefit of this description, will appreciate that other embodiments can be devised which do not depart from the scope of the invention as described herein. Furthermore, it should be noted that the language used in the specification has been principally selected for readability and instructional purposes, and may not have been selected to delineate or circumscribe the inventive subject matter. Accordingly, many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the appended claims. The present invention has been disclosed in an illustrative rather than a restrictive sense, and the scope of the present invention is defined by the appended claims.
Claims (16)
1. An image blur detection method, adapted to be executed in a computing device, the method comprising the steps of:
receiving an image to be detected;
intercepting a light spot image from the image, wherein the light spot image comprises at least one light spot;
extracting the contour line of each light spot in the light spot image;
for each of the contour lines extracted,
calculating the Hausdorff distance from the contour line to the corresponding standard contour line;
judging whether the calculated Hausdorff distance is smaller than a distance threshold of the standard contour line, wherein the distance thresholds of the standard contour line and the standard contour line are obtained in advance based on a plurality of non-fuzzy light spot images; and
if the proportion of contour lines with the Hausdorff distance to the corresponding standard contour line not less than the distance threshold of the standard contour line in all the extracted contour lines exceeds a preset proportion, determining that the image has blur, and determining that the blur type is motion blur; the method for obtaining the standard contour line and the distance threshold of the standard contour line based on the plurality of blur-free spot images comprises the following steps:
extracting the contour line of each light spot in each non-fuzzy light spot image;
dividing the extracted plurality of contour lines into at least one contour line set based on the Hausdorff distance;
for each set of contour lines,
in the contour line set, the step of selecting one alternative standard contour line is repeated until the selected alternative standard contour lines reach a preset number;
merging the selected candidate standard contour lines with a preset number into a standard contour line representing the contour line set, and calculating the distance threshold of the standard contour line based on the contour line set; wherein the step of selecting one of the candidate standard contour lines comprises: and for each contour line which is not selected as the alternative standard contour line in the contour line set, calculating the sum of the Hausdorff distances from the contour line which is not selected as the alternative standard contour line except the contour line to the contour line and all the alternative standard contour lines, and selecting the contour line with the minimum sum of the Hausdorff distances as the alternative standard contour line.
2. The image blur detection method according to claim 1, wherein the step of intercepting the spot image from the image comprises:
intercepting a region of interest from the image, the region of interest containing the spot image;
acquiring a light spot characteristic of the at least one light spot based on the region of interest;
intercepting the spot image according to the spot characteristics of the at least one spot.
3. The image blur detection method according to claim 2, wherein the step of acquiring the spot feature of the at least one spot based on the region of interest comprises:
carrying out threshold segmentation on the region of interest to obtain a binary image of the region of interest;
and determining the spot characteristics of the at least one spot according to the binary image.
4. The image blur detection method according to claim 3, wherein the spot feature of the at least one light spot includes a minimum bounding rectangle surrounding the at least one light spot, and the step of intercepting the spot image based on the spot feature of the at least one light spot includes:
intercepting the spot image according to a minimum bounding rectangle that encompasses the at least one spot.
5. The image blur detection method according to claim 2, wherein the spot characteristics of the at least one spot further include a minimum circumscribed rectangle aspect ratio and an area of each spot, the method further comprising the steps of:
after the light spot characteristics of at least one light spot are obtained based on the region of interest, whether the length-width ratio of the minimum circumscribed rectangle of each light spot meets a first light spot screening condition is judged;
if the proportion of the light spots of which the minimum external rectangular length-width ratio does not meet the first light spot screening condition in all the light spots exceeds a first light spot proportion, determining that the image is fuzzy, and determining that the fuzzy type is motion fuzzy;
if the ratio of the minimum length-width ratio of the circumscribed rectangle to the light spots meeting the first light spot screening condition in all the light spots exceeds the first light spot ratio, judging whether the area of each light spot meets the second light spot screening condition or not;
and if the proportion of the light spots with the areas not meeting the second light spot screening condition in all the light spots exceeds the second light spot proportion, determining that the image is fuzzy, and determining that the fuzzy type is focus fuzzy.
6. The image blur detection method according to claim 1, wherein the method further comprises the steps of:
after extracting the contour line of each light spot in the light spot image, acquiring the contour line characteristics of each contour line, wherein the contour line characteristics of each contour line comprise the perimeter, the area and the length-width ratio of the minimum circumscribed rectangle of the contour line;
for each contour line, judging whether the perimeter of the contour line meets a first contour line screening condition, whether the area meets a second contour line screening condition and whether the length-width ratio of the minimum circumscribed rectangle meets a third contour line screening condition;
and if the proportion of the contour line which does not meet any one of the first contour line screening condition, the second contour line screening condition and the third contour line screening condition in all the extracted contour lines exceeds the contour line proportion, determining that the image has blur, and determining that the type of the blur is focus blur.
7. The image blur detection method according to claim 1, wherein the method further comprises the steps of:
for each extracted contour line, respectively calculating the Hausdorff distance from the contour line to each standard contour line before calculating the Hausdorff distance from the contour line to the corresponding standard contour line;
and selecting the standard contour line with the minimum Hausdorff distance as the standard contour line corresponding to the contour line.
8. The image blur detection method according to claim 1, wherein the step of dividing the extracted plurality of contour lines into at least one set of contour lines comprises:
calculating the Hausdorff distance between every two of the extracted contour lines;
detecting whether contour lines which are not divided into any contour line set exist in the extracted contour lines;
if so, constructing a contour line set in contour lines which are not divided into any contour line set according to the calculated Hausdorff distance;
and repeating the steps of detecting the contour lines and constructing the contour line sets until no contour lines which are not divided into any contour line set exist in the plurality of extracted contour lines.
9. The image blur detection method according to claim 8, wherein the step of constructing one contour line set among contour lines which are not classified into any contour line set based on the calculated hausdorff distance includes:
selecting two contour lines with the minimum Hausdorff distance between every two contour lines in contour lines which are not added into any contour line set, and forming a contour line set;
in the contour lines which are not added into any contour line set, whether a contour line with the Hausdorff distance smaller than a preset threshold exists in any contour line in the newly formed contour line set is searched;
if the contour line set exists, respectively calculating the sum of the Hausdorff distances from each searched contour line to all contour lines in the newly formed contour line set, and adding the contour line with the minimum sum of the Hausdorff distances to the newly formed contour line set;
and repeating the steps of searching the contour lines, calculating the sum of the Hausdorff distances and adding the contour lines until no contour line with the Hausdorff distance smaller than a preset threshold value to any contour line in the newly formed contour line set exists in the contour lines without any contour line set.
10. The image blur detection method according to claim 9, wherein the predetermined threshold value is 0.15 to 0.19.
11. The image blur detection method of claim 1, wherein the step of calculating the distance threshold of the standard contour line based on the set of contour lines comprises:
respectively calculating the Hausdorff distance from each contour line in the contour line set to the standard contour line;
the maximum Housdov distance is selected as the distance threshold for the standard contour line.
12. The image blur detection method according to any one of claims 1 to 11, wherein the hausdorff distance is calculated by the following formula:
HD(A,B)＝max{h(A,B),h(B,A)}
wherein
A＝{pa1,pa2,pa3,......,pam}B＝{pb1,pb2,pb3,......,pbn}
HD (A, B) is the Hausdorff distance between A and B, h (A, B) and h (A, B) are the unidirectional Hausdorff distance from A to B and the unidirectional Hausdorff distance from B to A, A and B are two sets containing finite pixel points, | pa-pbI is pixel point paAnd pixel point pbEuclidean distance between, | | pb-paI is pixel point pbAnd pixel point paThe euclidean distance between.
13. The image blur detection method according to claim 1, wherein the step of extracting the contour line of the spot includes:
the isocline algorithm is used to extract the contour of the spot.
14. The image blur detection method according to claim 1, wherein the image is a near infrared image of the periocular or iris or human face.
15. A computing device, comprising:
one or more processors;
a memory; and
one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs comprising instructions for performing the image blur detection method of any of claims 1-14.
16. A readable storage medium storing a program, the program comprising instructions that, when executed by a computing device, cause the computing device to perform the image blur detection method according to any one of claims 1-14.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810497616.2A CN108665459B (en) | 2018-05-22 | 2018-05-22 | Image blur detection method, computing device and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810497616.2A CN108665459B (en) | 2018-05-22 | 2018-05-22 | Image blur detection method, computing device and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108665459A CN108665459A (en) | 2018-10-16 |
| CN108665459B true CN108665459B (en) | 2021-09-14 |
Family
ID=63776468
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810497616.2A Active CN108665459B (en) | 2018-05-22 | 2018-05-22 | Image blur detection method, computing device and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108665459B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110032315B (en) * | 2019-03-28 | 2021-04-27 | 广州柏视数据科技有限公司 | Rapid modification method for contour line of radiotherapy target area |
| CN112102147B (en) * | 2019-06-18 | 2022-03-08 | 腾讯科技（深圳）有限公司 | Background blurring identification method, device, equipment and storage medium |
| CN112329522A (en) * | 2020-09-24 | 2021-02-05 | 上海品览数据科技有限公司 | Goods shelf goods fuzzy detection method based on deep learning and image processing |
| CN114693954B (en) * | 2022-04-02 | 2024-07-09 | 中煤(天津)地下工程智能研究院有限公司 | Underground coal mine light spot feature extraction method and system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150172547A1 (en) * | 2013-12-13 | 2015-06-18 | Adobe Systems Incorporated | Image deblurring based on light streaks |
| CN105160306A (en) * | 2015-08-11 | 2015-12-16 | 北京天诚盛业科技有限公司 | Iris image blurring determination method and device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106126919A (en) * | 2016-06-23 | 2016-11-16 | 武汉大学 | A kind of accurate Hausdorff distance computational methods between any type point set data |
-
2018
- 2018-05-22 CN CN201810497616.2A patent/CN108665459B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150172547A1 (en) * | 2013-12-13 | 2015-06-18 | Adobe Systems Incorporated | Image deblurring based on light streaks |
| CN105160306A (en) * | 2015-08-11 | 2015-12-16 | 北京天诚盛业科技有限公司 | Iris image blurring determination method and device |
Non-Patent Citations (5)
| Title |
|---|
| A Real-Time Focusing Algorithm for Iris Recognition Camera;Park等;《IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews)》;20050725;第441-444页 * |
| 基于Hausdorff距离的轮廓线匹配;王文成;《中国优秀硕士学位论文全文数据库 信息科技辑（月刊） 计算机软件及计算机应用》;20070815(第02期);第I138-685页 * |
| 基于图像的激光光斑识别和特性分析;王亚丽;《中国优秀硕士学位论文全文数据库 信息科技辑（月刊） 计算机软件及计算机应用》;20090115(第01期);第I138-844页 * |
| 聚类评估算法-轮廓系数;山坡坡上的蜗牛;《https://blog.csdn.net/wangxiaopeng0329/article/details/53542606》;20161209;第1页 * |
| 虹膜图像预处理-运动模糊检测;continueOo;《https://blog.csdn.net/continueOo/article/details/71308280》;20170507;第1-2页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108665459A (en) | 2018-10-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108665459B (en) | Image blur detection method, computing device and readable storage medium | |
| CN106547744B (en) | Image retrieval method and system | |
| Zhang et al. | Image segmentation based on 2D Otsu method with histogram analysis | |
| JP6255486B2 (en) | Method and system for information recognition | |
| US9076056B2 (en) | Text detection in natural images | |
| CN110555372A (en) | Data entry method, device, equipment and storage medium | |
| CN109492642B (en) | License plate recognition method, license plate recognition device, computer equipment and storage medium | |
| US8842889B1 (en) | System and method for automatic face recognition | |
| CN110427946B (en) | Document image binarization method and device and computing equipment | |
| JP2002342756A (en) | Method for detecting position of eye and mouth in digital image | |
| WO2020082731A1 (en) | Electronic device, credential recognition method and storage medium | |
| US9058655B2 (en) | Region of interest based image registration | |
| WO2021184718A1 (en) | Card border recognition method, apparatus and device, and computer storage medium | |
| CN109447117B (en) | Double-layer license plate recognition method and device, computer equipment and storage medium | |
| CN113158773B (en) | Training method and training device for living body detection model | |
| US8913836B1 (en) | Method and system for correcting projective distortions using eigenpoints | |
| CN111898610A (en) | Card unfilled corner detection method and device, computer equipment and storage medium | |
| CN111222409A (en) | Vehicle brand labeling method, device and system | |
| CN108229583B (en) | Method and device for fast template matching based on main direction difference characteristics | |
| CN111047496A (en) | Threshold determination method, watermark detection device and electronic equipment | |
| CN111178200A (en) | Identification method of instrument panel indicator lamp and computing equipment | |
| US10373329B2 (en) | Information processing apparatus, information processing method and storage medium for determining an image to be subjected to a character recognition processing | |
| CN112532884A (en) | Identification method and device and electronic equipment | |
| KR102230559B1 (en) | Method and Apparatus for Creating Labeling Model with Data Programming | |
| CN111178340B (en) | Image recognition method and training method of image recognition model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20211210 Address after: 541000 building D2, HUTANG headquarters economic Park, Guimo Avenue, Qixing District, Guilin City, Guangxi Zhuang Autonomous Region Patentee after: Guangxi Code Interpretation Intelligent Information Technology Co.,Ltd. Address before: 201207 2 / F, building 13, 27 Xinjinqiao Road, Pudong New Area pilot Free Trade Zone, Shanghai Patentee before: SHIMA RONGHE (SHANGHAI) INFORMATION TECHNOLOGY Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230816 Address after: No. 8, Row 9, Fatou Dongli Middle Yard, Chaoyang District, Beijing, 100000 Patentee after: Wang Xiaopeng Address before: 541000 building D2, HUTANG headquarters economic Park, Guimo Avenue, Qixing District, Guilin City, Guangxi Zhuang Autonomous Region Patentee before: Guangxi Code Interpretation Intelligent Information Technology Co.,Ltd. |
|
| TR01 | Transfer of patent right |