WO2018095194A1 - 一种图像分析方法及装置 - Google Patents

一种图像分析方法及装置 Download PDF

Info

Publication number
WO2018095194A1
WO2018095194A1 PCT/CN2017/107872 CN2017107872W WO2018095194A1 WO 2018095194 A1 WO2018095194 A1 WO 2018095194A1 CN 2017107872 W CN2017107872 W CN 2017107872W WO 2018095194 A1 WO2018095194 A1 WO 2018095194A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
image
target
area
collection device
Prior art date
Application number
PCT/CN2017/107872
Other languages
English (en)
French (fr)
Inventor
谢忠贤
浦世亮
周明耀
Original Assignee
杭州海康威视数字技术股份有限公司
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 杭州海康威视数字技术股份有限公司 filed Critical 杭州海康威视数字技术股份有限公司
Priority to US16/463,951 priority Critical patent/US11048950B2/en
Priority to EP17873492.7A priority patent/EP3547253B1/en
Publication of WO2018095194A1 publication Critical patent/WO2018095194A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/0002Inspection of images, e.g. flaw detection
    • G06T7/0004Industrial image inspection
    • G06T7/001Industrial image inspection using an image reference approach
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/50Context or environment of the image
    • G06V20/52Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G06V20/54Surveillance or monitoring of activities, e.g. for recognising suspicious objects of traffic, e.g. cars on the road, trains or boats
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/70Determining position or orientation of objects or cameras
    • G06T7/73Determining position or orientation of objects or cameras using feature-based methods
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/50Context or environment of the image
    • G06V20/56Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • G06V20/58Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
    • G06V20/584Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/60Type of objects
    • G06V20/62Text, e.g. of license plates, overlay texts or captions on TV images
    • G06V20/63Scene text, e.g. street names
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30248Vehicle exterior or interior
    • G06T2207/30252Vehicle exterior; Vicinity of vehicle
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/60Type of objects
    • G06V20/62Text, e.g. of license plates, overlay texts or captions on TV images
    • G06V20/625License plates

Definitions

  • the present application relates to the field of image processing technologies, and in particular, to an image analysis method and apparatus.
  • image acquisition equipment In the field of traffic monitoring, image acquisition equipment is usually installed at intersections, bayonet, etc. to obtain an image containing the vehicle, and then relevant vehicle information is acquired based on the image.
  • the image capturing device may detect the collection area thereof, and when it is detected that the vehicle passes, capture the monitoring image including the target vehicle.
  • the target vehicle may be a vehicle detected by the image acquisition device.
  • the image acquisition device After the image acquisition device collects the image containing the target vehicle, in order to obtain more information of the target vehicle, such as vehicle brand information, attribution, etc., the image acquisition device can recognize the license plate information of the target vehicle and identify the license plate of the target vehicle. The information is sent to the image analysis system for analysis along with the corresponding image.
  • images captured by an image acquisition device may contain multiple vehicles, but only one target vehicle needs to be analyzed. Therefore, after receiving the license plate information and the corresponding image sent by the image acquisition device, the image analysis system may identify the target vehicle included in the image according to the license plate information, and then analyze the identified target vehicle.
  • the image analysis system needs to first analyze the entire image to identify the target vehicle contained therein. This will take a long time, resulting in low image analysis efficiency.
  • An object of the embodiments of the present application is to provide an image analysis method and apparatus to improve image analysis efficiency.
  • the specific technical solutions are as follows:
  • an embodiment of the present application provides an image analysis method, where the method includes:
  • the target image collection device And receiving, by the target image collection device, the target information, where the target information includes at least: the first image collected by the target image collection device, and the identification information of the target image collection device;
  • a second image corresponding to the target area information is determined, and the second image is analyzed.
  • the target information further includes: first license plate information of the target vehicle included in the first image; and the analyzing the second image includes:
  • the method further includes:
  • analyzing the first image determining first region information of the first image, and correspondingly storing identifier information of the target image capturing device, first region information of the first image, and a second determination result, wherein the second determination result is that the first license plate information is different from the second license plate information.
  • the step of determining the first area information of the second image includes:
  • the determining the first area information of the first image includes:
  • the method further includes:
  • the steps of the area information corresponding to the identification information of the device include:
  • Determining the second region information based on a statistical result of the first region information of the plurality of first images, or based on a feature matching result of the first region information of the plurality of first images;
  • the target image collection is updated according to the identifier information of the target image collection device and the first region information of the second image corresponding to each first determination result saved in the third preset time period.
  • the steps of the area information corresponding to the identification information of the device include:
  • Determining the second region information based on a statistical result of the first region information of the plurality of second images, or based on a feature matching result of the first region information of the plurality of second images;
  • the step of pre-storing the correspondence between the identification information of each image collection device and each area information includes:
  • the area information corresponding to the image collection device is determined and saved according to the initial area information corresponding to the identification information of the image collection device.
  • an embodiment of the present application provides an image analysis apparatus, where the apparatus includes:
  • a first receiving module configured to receive target information that is sent by the target image capturing device, where the target information includes at least: a first image collected by the target image capturing device, and identification information of the target image capturing device;
  • An execution module configured to acquire target area information corresponding to the identification information of the target image collection device according to the correspondence between the identification information of each image collection device and the area information saved in advance;
  • an analysis module configured to determine, in the first image, a second image corresponding to the target area information, and analyze the second image.
  • the target information further includes: first license plate information of the target vehicle included in the first image; the analyzing module is specifically configured to identify second license plate information in the second image;
  • the device also includes:
  • a first determining module configured to determine whether the first license plate information and the second license plate information are the same
  • a first storage module configured to: when the first determining module determines that the result is yes, analyze the second image, determine first region information of the second image, and save the target image capturing device correspondingly Identification information, first region information of the second image, and first determination result, wherein the first determination result is that the first license plate information is the same as the second license plate information;
  • a second storage module configured to: when the first determining module determines that the result is negative, analyze the first image, determine first region information of the first image, and save the target image capturing device correspondingly Identification information, first region information of the first image, and second determination result, wherein the second determination result is that the first license plate information is different from the second license plate information.
  • the first storage module includes:
  • a first identification submodule configured to identify a target vehicle in which the license plate information in the second image is the first license plate information
  • the first determining submodule is configured to determine location information of an area occupied by the target vehicle in the second image, and determine the location information as first area information of the second image.
  • the second storage module includes:
  • a second identification submodule configured to identify a target vehicle in which the license plate information in the first image is the first license plate information
  • a second determining submodule configured to determine location information of an area occupied by the target vehicle in the first image, and determine the location information as first area information of the first image.
  • the device further includes:
  • a second determining module configured to determine, according to the set time interval, whether the number of the second determination results saved in the first preset time period is greater than a first preset threshold
  • a first update module configured to: when the second determination module determines that the result is yes, the identification information of the target image collection device corresponding to each second determination result saved in the second preset time period Information, the first area information of the first image, and the area information corresponding to the identification information of the target image collection device;
  • a second update module configured to: when the second determination module determines that the result is no, the identification information of the target image collection device corresponding to each first determination result saved in the third preset time period, the first The first area information of the two images is updated, and the area information corresponding to the identification information of the target image collection device is updated.
  • the first update module includes:
  • a first acquiring sub-module configured to acquire first region information of the plurality of first images corresponding to the identifier information of the target image capturing device from each of the second determining results saved in the second preset time period;
  • a third determining submodule configured to determine second region information based on a statistical result of the first region information of the plurality of first images, or based on a feature matching result of the first region information of the plurality of first images;
  • the first update submodule is configured to update the area information corresponding to the identifier information of the target image collection device according to the determined second area information.
  • the second update module includes:
  • a second acquiring sub-module configured to acquire, from each of the first determination results saved in the third preset time period, first region information of the plurality of second images corresponding to the identifier information of the target image capturing device;
  • a fourth determining submodule configured to determine second region information based on a statistical result of the first region information of the plurality of second images, or based on a feature matching result of the first region information of the plurality of second images;
  • a second update submodule configured to update area information corresponding to the identifier information of the target image collection device according to the determined second area information.
  • the device further includes:
  • a second receiving module configured to receive, for each image capturing device, a third image sent by the image capturing device, third license plate information of a target vehicle included in the third image, and the image capturing Collecting identification information of the device;
  • a processing module configured to identify a target vehicle included in the third image as the target vehicle of the third license plate information, and determine location information of an area occupied by the target vehicle in the third image, the location information Determining as initial region information of the third image;
  • a third storage module configured to save a correspondence between the identification information of the image collection device and the initial region information
  • a fourth storage module configured to: when the number of initial area information corresponding to the image collection device that is saved by the third storage module is greater than a second preset threshold, according to each initial area information corresponding to the identification information of the image collection device, The area information corresponding to the image collection device is determined and saved.
  • the application provides an electronic device, including:
  • the memory stores executable program code
  • the processor runs a program corresponding to the executable program code by reading executable program code stored in the memory for performing an image analysis method according to the first aspect of the present application at runtime .
  • the embodiment of the present application provides an image analysis method and apparatus, and the method includes: receiving target information sent by a target image collection device; wherein the target information includes at least: a first image collected by the target image acquisition device, And the identification information of the target image collection device; the target area information corresponding to the identification information of the target image collection device is acquired according to the correspondence between the identification information of each image collection device and the regional information stored in advance; In the first image, a second image corresponding to the target area information is determined, and the second image is analyzed.
  • the area where the target vehicle is located is generally relatively fixed. Therefore, the corresponding relationship between the identification information of each image capturing device and each area information can be obtained in advance.
  • the information includes location information of an area in which the target vehicle is located in the image acquired by each image acquisition device.
  • the second image corresponding to the target region information may be determined according to the target region information corresponding to the target image capturing device, and only the second image is analyzed, compared with the first image. , the size of the second image Smaller, therefore, can improve image analysis efficiency.
  • FIG. 1 is a schematic diagram of an image acquisition area and a detection area of an image acquisition device
  • FIG. 2 is a flowchart of an image analysis method according to an embodiment of the present application.
  • 3(a) is a schematic diagram of a first image acquired by a target image acquisition device according to an embodiment of the present application
  • 3(b) is a second image diagram of the first image shown in FIG. 3(a) and the target area information;
  • FIG. 4 is another flowchart of an image analysis method according to an embodiment of the present application.
  • FIG. 5 is another flowchart of an image analysis method according to an embodiment of the present application.
  • FIG. 6 is a schematic diagram of determining area information according to each initial area information according to an embodiment of the present application.
  • FIG. 7 is another flowchart of an image analysis method according to an embodiment of the present application.
  • FIG. 8 is a schematic structural diagram of an image analysis apparatus according to an embodiment of the present disclosure.
  • FIG. 9 is another schematic structural diagram of an image analysis apparatus according to an embodiment of the present disclosure.
  • FIG. 10 is another schematic structural diagram of an image analysis apparatus according to an embodiment of the present application.
  • FIG. 11 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
  • an embodiment of the present application provides an image analysis method and apparatus.
  • an image acquisition device installed at an intersection, a bayonet, or the like can detect an image acquisition area thereof, and when it is detected that a vehicle passes, captures a surveillance image including the target vehicle. Moreover, in general, for each image acquisition device, when its installation is completed, its position, acquisition angle of view and other parameters are determined, and therefore, the image acquisition area thereof is relatively fixed.
  • the image capturing device does not perform vehicle detection in the entire image capturing area when performing vehicle detection, but in a relatively small detection area that is relatively fixed and included in the collection area thereof.
  • Vehicle detection Only when there is a vehicle passing through the detection area of the image capturing device can it detect that there is a vehicle passing, and when the vehicle is in the image capturing area outside the detection area, the image collecting device cannot detect the vehicle. That is to say, the detection area is the area of interest of the image acquisition device, and is also the area where the target vehicle is located in the image collected by the image acquisition device.
  • the image collected by the image acquisition device also includes an image acquisition area and a detection area.
  • the image collection area may be an entire image area
  • the detection area is an image area corresponding to an actual detection area of the image collection device.
  • the detection area when analyzing the image collection area, the detection area may be used as a preferred area for vehicle detection, and the non-detection area within the image acquisition area may be used as a secondary area for vehicle detection.
  • the image collection area of the image acquisition device may be the area 110, and the detection area is the area 120.
  • the range of the image acquisition area 110 of the image acquisition device is generally large, while the range of the detection area 120 is relatively small.
  • the area information of the image captured by each image acquisition device may be determined in advance, that is, the position information of the detection area in the image collected by the image acquisition device is determined, and correspondingly The correspondence between the identification information of each image collection device and the area information is saved.
  • the image information may be determined by the image analysis system according to the image collected by each image acquisition device, and may be saved by another self-learning system according to the image collected by each image acquisition device.
  • the image analysis system can establish a connection relationship with the self-learning system.
  • the image analysis method provided by the image analysis system is determined according to the image collected by each image acquisition device, and the area information is determined and saved as an example.
  • the correspondence between the identification information of each image capturing device and the area information may be stored locally in the image analysis system, or may be stored in an external device that establishes a connection relationship with the image analysis system.
  • a coordinate system may be set for the image acquired by each image acquisition device. As shown in FIG. 1, one vertex of the image may be determined as the origin O, and determined. X axis, Y axis.
  • the start point and the end point in the X direction of the detection area, and the start point and the end point in the Y direction can be saved correspondingly.
  • the starting point and the ending point of the saved X direction, and the starting point and the ending point of the Y direction may be pixels.
  • the embodiment of the present application provides an image analysis method process. As shown in FIG. 2, the process may include the following steps:
  • the target information includes at least: a first image collected by the target image collection device and identifier information of the target image collection device.
  • the method provided by the embodiments of the present application can be applied to an image analysis system.
  • the image analysis system can operate in an electronic device such as a desktop computer, a portable computer, a smart mobile terminal, or the like.
  • an image acquisition device in order to perform traffic monitoring, may be installed at a junction, a bayonet, or the like where monitoring is required.
  • the image capturing device may be a ball machine, a camera, a camera, or the like, which is not limited in this embodiment of the present application.
  • a wired or wireless connection can be established between the image acquisition device and the image analysis system so that the image acquisition device can transmit its acquired image to the image analysis system.
  • the connection between the image acquisition device and the image analysis system can be established through a wireless connection method such as WIFI (Wireless Fidelity), NFC (Near Field Communication), Bluetooth, etc. This is not limited.
  • the method provided by the embodiment of the present application can analyze the image collected by each image acquisition device.
  • the image analysis method provided by the embodiment of the present application is described by using only one image acquisition device as an example. Also, for the convenience of description, the image acquisition device for the embodiment of the present application may be referred to as a target image acquisition device.
  • the target image capturing device may detect whether there is a vehicle passing through the detection area, and when detected, may collect a first image including the detected target vehicle. Moreover, the target image acquisition device can transmit the first image it collects and its own identification information to the image analysis system, so that the image analysis system acquires more information of the target vehicle.
  • the target collecting device may also collect an image of the detecting area when receiving the indication signal of the vehicle detecting device, for example, the target vehicle passes through the infrared ray emitted by the infrared sensing device.
  • the infrared sensing device sends an indication signal to the vehicle detecting device to indicate that the target collecting device collects an image, and when the target collecting device collects an image (such as a captured image of the capturing machine), the target vehicle is also located.
  • the target acquisition device can acquire the first image of the target vehicle.
  • an induction (such as a weight sensor) is disposed on the road surface near the detection area, and once the vehicle drives over the road surface, the sensing device instructs the target collection device (such as a camera) to capture video or images, and the target vehicle can be collected. An image.
  • the target collection device such as a camera
  • the image analysis system may receive the target information sent by the target image collection device, where the target information may include at least: the first image collected by the target image collection device and the identification information of the target image collection device.
  • the identification information of the target image collection device included in the target information received by the image analysis system may be: 02
  • the first image included in the target information received by the image analysis system may be as shown in FIG. 3(a).
  • the image analysis system may acquire the identifier of the target image collection device according to the correspondence between the identifier information of each image collection device and each region information saved in advance. Target area information corresponding to the information. Specifically, the image analysis system may search for the same identification information as the identification information of the target image collection device in the identification information of each image collection device, and determine the region information corresponding to the identified identification information as the target region information.
  • the identifier of the target image collection device acquired
  • the target area information corresponding to the information is: X (112, 232), Y (21, 53).
  • the image analysis system may determine a second image corresponding to the target area information in the first image, and perform the second image on the second image. analysis.
  • the image analysis system may determine an area corresponding to the target area information in the first image, and determine an image including the area as the second image, so that the second image may be analyzed.
  • the identified target area information is: X (112, 232), Y (21, 53)
  • the X-axis direction is from 112 pixels to 232 pixels
  • the Y-axis direction from 21 pixels to 53 pixels is an area corresponding to the target area information, as in the area 310 in FIG. 3(a).
  • the image containing the area is Second image.
  • the second image determined by the image acquisition device can be as shown in Figure 3(b).
  • the second image may be an image including a target vehicle, or may be an image including only a license plate.
  • information such as the vehicle brand and the attribution of the target vehicle included in the second image may be identified, which is not limited in this embodiment of the present application.
  • the area where the target vehicle is located is generally relatively fixed. Therefore, the corresponding relationship between the identification information of each image capturing device and each area information can be obtained in advance.
  • the information includes location information of an area in which the target vehicle is located in the image acquired by each image acquisition device.
  • the second image corresponding to the target region information may be determined according to the target region information corresponding to the target image capturing device, and only the second image is analyzed, compared with the first image. The size of the second image is small, and therefore, the image analysis efficiency can be improved.
  • the image analysis system pre-stores the correspondence between the identification information of the target image collection device and the target region information
  • the target region information is information of the vehicle
  • the target region information corresponds to the second
  • the vehicle included in the image is larger in size, and the vehicle of the same size does not appear again for a long period of time, which will result in the target area information corresponding to the saved identification information of the target image capturing device being inaccurate; or
  • the target area information is information of the license plate
  • the target area information corresponds to the second image that only the license plate does not contain the entire vehicle, and the size of the license plate is relatively fixed, but when the target image acquisition device is due to some external factors
  • the target area information corresponding to the target image acquisition device stored in advance may not be accurate.
  • the image analysis system can optimize and update the area information corresponding to the target image collection device to ensure the accuracy of the area information corresponding to the target image collection device, so as to further improve the image analysis efficiency.
  • the image analysis method provided by the embodiment of the present application may include the following steps:
  • S201 Receive target information sent by the target image collection device, where the target information includes at least: a first image collected by the target image collection device, identification information of the target image collection device, and the first image.
  • the target image capturing device when the target image capturing device detects that the detection area has the target vehicle, it can identify the license plate information of the target vehicle and send it to the image analysis system, so that the image analysis system identifies the corresponding information according to the license plate information.
  • the image analysis system identifies the corresponding information according to the license plate information.
  • Target vehicle and analyze the target vehicle.
  • the image capture device can identify the license plate information of the target vehicle included in the first image by using any image recognition method, which is not described in this embodiment of the present application.
  • the target information received by the image analysis system may include: a first image collected by the target image capturing device, identification information of the target image capturing device, and a target vehicle included in the first image.
  • a license plate information may be included in the first image.
  • This step is basically the same as step S102 in the embodiment shown in FIG. 2, and details are not described herein.
  • S203 Determine, in the first image, a second image corresponding to the target area information, and identify second license plate information in the second image.
  • At least the second license plate information in the second image may be identified, according to the second license plate information, and the target vehicle sent by the target image capturing device.
  • the image analysis system After the image analysis system recognizes the second license plate information of the vehicle included in the second image, it can determine whether the first license plate information sent by the target image collection device is the same as the second license plate information.
  • the image analysis system determines that the first license plate information is the same as the second license plate information, it indicates that the area information corresponding to the pre-stored target image acquisition device is correct.
  • the detection area corresponding to the area information corresponding to the target image collection device saved in advance It may be larger than the actual detection area of the target image acquisition device. Therefore, the image analysis system can analyze the second image to determine the first region information of the second image.
  • the first area information of the second image may be the location information of the area where the vehicle is located in the second image, or may be the location information of the area where the license plate is located in the second image.
  • the image analysis system may identify the target vehicle in which the license plate information in the second image is the first license plate information; and determine position information of the area occupied by the target vehicle in the second image, and determine the position information as the first image of the second image.
  • Regional information may identify the image analysis system.
  • the image analysis system may identify the license plate information in the second image as the license plate of the first license plate information; and determine position information of the area occupied by the license plate in the second image, and determine the position information as the first area of the second image. information.
  • the format of the first area information saved by the image analysis system may be the same as the format of the area information saved in advance.
  • the image analysis system may correspond to the identification information of the target image collection device, the first area information of the second image, and the first determination result, wherein the first determination result is the first license plate
  • the information is the same as the second license plate information, so as to optimize and update the area information corresponding to the target image collection device according to the saved information.
  • the image analysis system determines that the first license plate information is different from the second license plate information, it indicates that the area information corresponding to the pre-stored target image collection device is not accurate enough.
  • the image analysis system may perform an overall analysis on the first image to determine the first region information of the first image.
  • the first area information of the first image may be the location information of the area where the vehicle is located in the first image, or may be the location information of the area where the license plate is located in the first image.
  • the image analysis system may identify the target vehicle in which the license plate information in the first image is the first license plate information; and determine the position information of the area occupied by the target vehicle in the first image, and determine the position information as the first image of the second image. Regional information.
  • the image analysis system may identify the license plate information in the first image as the license plate of the first license plate information; and determine position information of the area occupied by the license plate in the first image, and determine the position information as the first area of the first image. information.
  • the image analysis system may correspond to the identification information of the target image collection device, the first area information of the first image, and the second determination result, wherein the second determination result is the first license plate
  • the information is different from the second license plate information, so as to optimize and update the area information corresponding to the target image capturing device according to the saved information.
  • the image analysis system may obtain the first license plate information of the target vehicle transmitted by the target image acquisition device, and identify the second license plate information in the second image, and when the first license plate information and the second license plate information are the same, determine And saving the first area information of the second image; when the first license plate information is different from the second license plate information, determining and saving the first area information of the first image; thereby, corresponding to the target image collecting device according to the saved information
  • the area information is optimized for updates.
  • the image analysis system when the image analysis system predetermines and saves the correspondence between each image collection device and each region information, it may periodically update the region information corresponding to each image collection device.
  • the self-learning system when the self-learning system predetermines and saves the correspondence between each image capturing device and each region information, it can also periodically update the region information corresponding to each image capturing device in a similar manner.
  • the image analysis method provided by the embodiment of the present application, as shown in FIG. 5, may further include:
  • step S207 Determine, according to the set time interval, whether the number of the second determination results saved in the first preset time period is greater than a first preset threshold; if yes, execute step S208; if no, perform step S209.
  • the image analysis system may determine, according to the set time interval, such as 2 minutes, 5 minutes, 10 minutes, etc., whether the number of second determination results saved in the first preset time period is greater than the first Preset threshold.
  • the first preset time period may be, for example, 2 hours, 5 hours, 12 hours, etc.
  • the first preset threshold may be, for example, 100, 200, 300, or the like.
  • the image analysis system determines the number of second determination results saved in the first preset time period When the value is greater than the first preset threshold, it indicates that the first license plate information is not recognized from the second image in the first preset time period, so that the area information corresponding to the target image collection device may be determined to be less accurate. , or has changed, in this case, the area information corresponding to the saved target image collection device can be updated.
  • the image analysis system may acquire, from each of the second determination results saved in the second preset time period, the first region information of the plurality of first images corresponding to the identifier information of the target image collection device, and then based on the plurality of a statistical result of the first region information of an image, or a feature matching result based on the first region information of the plurality of first images, determining the second region information, and finally updating the target image capturing device according to the determined second region information
  • the second preset time period and the first preset time period may be the same or different.
  • the image analysis system may determine, according to the statistical result of the first region information of the plurality of first images, the process of determining the second region information, for example, the first region corresponding to each first region information may be first determined. A second area covering each of the first areas is then determined, and the location information of the second area is determined as the second area information. As shown in FIG. 6 , when the first regions corresponding to the plurality of first region information are the regions 610 , 620 , and 630 , respectively, the second region corresponding to the determined second region information may be the region 640 .
  • the image analysis system may also collect, in the plurality of first area information, the first area information with the highest probability of occurrence, and determine the first area information as the second area information.
  • the image analysis system may further determine a second region having the highest overlap rate in the first region corresponding to the plurality of first region information, and determine location information of the second region as the second region information.
  • the image analysis system may determine a center point of the first area corresponding to each of the first area information, and then expand to a second area of the preset license plate size centering on the center point, and determine location information of the second area as Second area information.
  • the image analysis system determines the process of the second region information based on the feature matching result of the first region information of the plurality of first images, which may be a Scale Invariant Feature Transform (SIFT)
  • SIFT Scale Invariant Feature Transform
  • the image analysis system determines that the number of the second determination results saved in the first preset time period is not greater than the first preset threshold, indicating that the first image is not recognized in the first preset time period.
  • the number of license plate information is small, so that it can be determined that the area information corresponding to the target image capturing device is relatively accurate.
  • the image analysis system can further optimize the region information corresponding to the target image acquisition device.
  • the image analysis system may acquire, from each of the first determination results saved in the third preset time period, the first region information of the plurality of second images corresponding to the identification information of the target image collection device, and then based on the plurality of second Determining the second region information based on the statistical result of the first region information of the image or the feature matching result of the first region information of the plurality of second images, and finally updating the identifier of the target image capturing device according to the determined second region information
  • the third preset time period and the first preset time period may be the same or different.
  • the image analysis system may determine, according to the statistical result of the first region information of the plurality of second images, the process of determining the second region information, for example, the first region corresponding to each first region information may be first determined. A second area covering each of the first areas is then determined, and the location information of the second area is determined as the second area information.
  • the image analysis system may also collect, in the plurality of first area information, the first area information with the highest probability of occurrence, and determine the first area information as the second area information.
  • the image analysis system may further determine a second region having the highest overlap rate in the first region corresponding to the plurality of first region information, and determine location information of the second region as the second region information.
  • the image analysis system may determine a center point of the first area corresponding to each of the first area information, and then expand to a second area of the preset license plate size centering on the center point, and determine location information of the second area as Second area information.
  • the image analysis system may perform feature matching on the first region information of the plurality of second images based on the SIFT algorithm, and use the feature matching result as the second region information.
  • the image analysis system may optimize and update the saved region information corresponding to each image collection device according to the saved first region information of the first image or the first region information of the second image, thereby ensuring each image.
  • the accuracy of the area information corresponding to the device is collected, thereby improving the efficiency of image analysis.
  • the image analysis system may pre-store the correspondence between each image collection device and each area information. Specifically, as shown in FIG. 7, the image analysis method provided by the embodiment of the present application may further include the following steps:
  • each image collection device receives a third image sent by the image collection device, third license plate information of the target vehicle included in the third image, and identification information of the image collection device.
  • the image analysis system may receive, for each image acquisition device, a third image sent by the image acquisition device, third license plate information of the target vehicle included in the third image, and identification information of the image collection device. .
  • the third license plate information is identified by the image acquisition device from the third image.
  • the image capture device may detect the target vehicle in the third image corresponding region according to its detection region and identify the third license plate information of the target vehicle.
  • the image analyzing system may identify the target vehicle included in the third image as the third license plate information.
  • the image analysis system may employ any image recognition method in which the license plate information is identified as the vehicle of the third license plate information, and the recognized license plate is determined as the target vehicle.
  • the image analysis system may further determine location information of the area occupied by the target vehicle in the third image. For example, the image analysis system may establish a coordinate system in the third image, thereby determining a starting point, an ending point, and a starting point and an ending point in the Y direction of the area occupied by the target vehicle in the third image, and determining the position. The information is determined as the initial area information of the third image.
  • the image analysis system may save the correspondence between the identification information of the image collection device and the initial region information. For example, the image analysis system may save the identification information of the image collection device and the initial region information locally, or may save it in itself Establish an external device in the connection relationship.
  • the image analysis system may store, for each image collection device, initial region information corresponding to the identification information of each image collection device.
  • initial region information corresponding to the identification information of each image collection device.
  • Table 2 the correspondence between the identification information of each image collection device and the initial region information saved by the image analysis system can be as shown in Table 2:
  • the image analysis system may determine the initial area information corresponding to the identification information of the image collection device.
  • the area information corresponding to the image collection device is saved. Specifically, the image analysis system may determine the area information corresponding to the image collection device based on the statistical result or the feature matching result of the plurality of initial area information corresponding to the image collection device.
  • the process of determining, by the image analysis system, the region information corresponding to the image collection device based on the statistics of the plurality of initial region information corresponding to the image collection device may be, for example, determining an initial corresponding to each initial region information. Area, then determine to cover each initial area The target area of the domain, and the location information of the target area is determined as the area information corresponding to the image collection device.
  • the correspondence between the identification information of each image collection device and the initial region information may be as shown in Table 3:
  • the image analysis system may also collect initial region information with the highest probability of occurrence in the plurality of initial region information, and determine the initial region information as the region information corresponding to the image capturing device.
  • the image analysis system may determine the target area with the highest overlap rate in the initial area corresponding to the plurality of initial area information, and determine the location information of the target area as the area information corresponding to the image collection device.
  • the image analysis system may determine a center point of the initial region corresponding to each initial region information, and then expand to a target region of a preset license plate size centering on the center point, and determine location information of the target region as the image capturing device. Corresponding area information.
  • the image analysis system may obtain the feature matching result of the plurality of initial region information corresponding to the image capturing device by using the SIFT algorithm, as the region information corresponding to the image capturing device.
  • the image analysis system may pre-store the correspondence between each image collection device and each region information, and further, when receiving the first image sent by the target image collection device, may determine to correspond to the region information of the target image collection device.
  • the second image is analyzed only for the second image, and the size of the second image is smaller than that of the first image, and therefore, image analysis efficiency can be improved.
  • the embodiment of the present application also provides a corresponding device embodiment.
  • FIG. 8 is an image analysis apparatus according to an embodiment of the present application, where the apparatus includes:
  • the first receiving module 810 is configured to receive target information that is sent by the target image capturing device, where the target information includes at least: a first image collected by the target image capturing device, and identifier information of the target image capturing device;
  • the execution module 820 is configured to acquire target area information corresponding to the identification information of the target image collection device according to the correspondence between the identification information of each image collection device and the area information saved in advance;
  • the analyzing module 830 is configured to determine, in the first image, a second image corresponding to the target area information, and analyze the second image.
  • the area where the target vehicle is located is generally relatively fixed. Therefore, the corresponding relationship between the identification information of each image capturing device and each area information can be obtained in advance.
  • the information may be location information of an area in which the target vehicle is located in the image acquired by each image acquisition device.
  • the second image corresponding to the target region information may be determined according to the target region information corresponding to the target image capturing device, and only the second image is analyzed, compared with the first image. The size of the second image is small, and therefore, the image analysis efficiency can be improved.
  • the target information further includes: first license plate information of the target vehicle included in the first image; the analyzing module is specifically configured to identify the second image Second license plate information; as shown in FIG. 9, the device further includes:
  • the first determining module 840 is configured to determine whether the first license plate information and the second license plate information are the same;
  • the first storage module 850 is configured to analyze the second image when the first determination module 840 determines that the result is YES, determine the first region information of the second image, and save the target image correspondingly And the first determination result that the first license plate information is the same as the second license plate information;
  • the second storage module 860 is configured to analyze the first image, determine the first area information of the first image, and save the target image correspondingly when the first determining module 840 determines that the result is no.
  • the first storage module 850 includes:
  • a first identification sub-module (not shown) for identifying a target vehicle in which the license plate information in the second image is the first license plate information
  • a first determining sub-module (not shown) for determining location information of an area occupied by the target vehicle in the second image, and determining the location information as a first area of the second image information.
  • the second storage module 860 includes:
  • a second identification sub-module (not shown) for identifying a target vehicle in which the license plate information in the first image is the first license plate information
  • a second determining sub-module (not shown) for determining location information of an area occupied by the target vehicle in the first image, and determining the location information as a first area of the first image information.
  • the device further includes:
  • the second determining module 870 is configured to determine, according to the set time interval, whether the number of the second determination results saved in the first preset time period is greater than a first preset threshold;
  • the first update module 880 is configured to: when the second determination module 870 determines that the result is YES, the identifier information and the identifier of the target image collection device corresponding to each second determination result saved in the second preset time period.
  • the first area information of the first image is updated, and the area information corresponding to the identification information of the target image collection device is updated;
  • the second update module 890 is configured to: when the second determination module 870 determines that the result is no, the identification information and the identifier of the target image collection device corresponding to each first determination result saved in the third preset time period Decoding the first area information of the second image, updating the identifier of the target image collection device The area information corresponding to the information.
  • the first update module 880 includes:
  • a first acquiring sub-module (not shown), configured to acquire, according to each second determination result saved in the second preset time period, a plurality of first corresponding to the identifier information of the target image capturing device First area information of the image;
  • a third determining submodule for using a statistical result of the first region information of the plurality of first images or a feature matching result based on the first region information of the plurality of first images Determining the second area information;
  • the first update sub-module (not shown) is configured to update the area information corresponding to the identifier information of the target image collection device according to the determined second area information.
  • the second update module 890 includes:
  • a second acquiring sub-module (not shown), configured to acquire, from each of the first determination results saved in the third preset time period, a plurality of second corresponding to the identification information of the target image capturing device First area information of the image;
  • a fourth determining submodule for using a statistical result of the first region information of the plurality of second images or a feature matching result based on the first region information of the plurality of second images Determining the second area information;
  • the second update sub-module (not shown) is configured to update the area information corresponding to the identifier information of the target image collection device according to the determined second area information.
  • the device further includes:
  • the second receiving module 1010 is configured to receive, for each image capturing device, a third image sent by the image capturing device, third license plate information of the target vehicle included in the third image, and identification information of the image capturing device;
  • a processing module 1020 configured to identify a target vehicle included in the third image as the target vehicle of the third license plate information, and determine location information of an area occupied by the target vehicle in the third image, the location Information is determined as initial region information of the third image;
  • the third storage module 1030 is configured to save a correspondence between the identifier information of the image collection device and the initial region information.
  • the fourth storage module 1040 is configured to: when the number of initial area information corresponding to the image collection device saved by the third storage module 1030 is greater than a second preset threshold, each initial region corresponding to the identifier information of the image collection device Information, determine and save the area information corresponding to the image collection device.
  • the image analysis efficiency can be improved.
  • the embodiment of the present application further provides an electronic device, which may include: a processor and a memory;
  • the memory stores executable program code
  • the processor runs a program corresponding to the executable program code by reading executable program code stored in the memory for executing the image analysis method described above.
  • the embodiment of the present application further provides an electronic device, which may include:
  • processor 1110 a processor 1110, a memory 1120, a communication interface 1130, and a bus 1140;
  • the processor 1110, the memory 1120, and the communication interface 1130 are connected through the bus 1140 and complete communication with each other;
  • the memory 1120 stores executable program code
  • the processor 1110 runs a program corresponding to the executable program code by reading executable program code stored in the memory 1120, for performing an image analysis according to an embodiment of the present application at runtime.
  • the method, wherein the image analysis method comprises:
  • the target image collection device And receiving, by the target image collection device, the target information, where the target information includes at least: the first image collected by the target image collection device, and the identification information of the target image collection device;
  • a second image corresponding to the target area information is determined, and the second image is analyzed.
  • the area where the target vehicle is located is generally relatively fixed. Therefore, the corresponding relationship between the identification information of each image capturing device and each area information can be obtained in advance.
  • the information includes location information of an area in which the target vehicle is located in the image acquired by each image acquisition device.
  • the second image corresponding to the target region information may be determined according to the target region information corresponding to the target image capturing device, and only the second image is analyzed, compared with the first image. The size of the second image is small, and therefore, the image analysis efficiency can be improved.
  • the embodiment of the present application further provides a storage medium, where the storage medium is used to store executable program code, and the executable program code is used to execute an image according to an embodiment of the present application at runtime.
  • An analysis method wherein the image analysis method comprises:
  • the target image collection device And receiving, by the target image collection device, the target information, where the target information includes at least: the first image collected by the target image collection device, and the identification information of the target image collection device;
  • a second image corresponding to the target area information is determined, and the second image is analyzed.
  • the area where the target vehicle is located is generally relatively fixed. Therefore, the corresponding relationship between the identification information of each image capturing device and each area information can be obtained in advance.
  • the information includes location information of an area in which the target vehicle is located in the image acquired by each image acquisition device.
  • the second image corresponding to the target region information may be determined according to the target region information corresponding to the target image capturing device, and only the second image is analyzed, compared with the first image. The size of the second image is small, and therefore, the image analysis efficiency can be improved.
  • the embodiment of the present application further provides an application program, where the application is used to execute an image analysis method according to an embodiment of the present application at runtime, where the image analysis is performed.
  • Methods include:
  • the target image collection device And receiving, by the target image collection device, the target information, where the target information includes at least: the first image collected by the target image collection device, and the identification information of the target image collection device;
  • a second image corresponding to the target area information is determined, and the second image is analyzed.
  • the area where the target vehicle is located is generally relatively fixed. Therefore, the corresponding relationship between the identification information of each image capturing device and each area information can be obtained in advance.
  • the information includes location information of an area in which the target vehicle is located in the image acquired by each image acquisition device.
  • the second image corresponding to the target region information may be determined according to the target region information corresponding to the target image capturing device, and only the second image is analyzed, compared with the first image. The size of the second image is small, and therefore, the image analysis efficiency can be improved.
  • the description is relatively simple, and the relevant parts can be referred to the description of the method embodiment.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Quality & Reliability (AREA)
  • Traffic Control Systems (AREA)

Abstract

一种图像分析方法及装置,所述方法包括:接收目标图像采集设备发送的目标信息;其中,所述目标信息至少包括:所述目标图像采集设备采集的第一图像、以及所述目标图像采集设备的标识信息(S101);根据预先保存的各图像采集设备的标识信息与各区域信息的对应关系,获取与所述目标图像采集设备的标识信息对应的目标区域信息(S102);在所述第一图像中,确定与所述目标区域信息对应的第二图像,并对所述第二图像进行分析(S103)。能够仅对包含第一图像特定区域的第二图像进行分析,与第一图像相比,第二图像的尺寸较小,因此,能够提高图像分析效率。

Description

一种图像分析方法及装置
本申请要求于2016年11月25日提交中国专利局、申请号为201611056740.2发明名称为“一种图像分析方法及装置”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及图像处理技术领域,特别是涉及一种图像分析方法及装置。
背景技术
在交通监控领域,通常会在路口、卡口等地方安装图像采集设备,以获得包含车辆的图像,进而根据该图像获取相关的车辆信息。具体地,图像采集设备可以对其采集区域进行检测,当检测到有车辆通过时,抓拍得到包含目标车辆的监控图像。其中,目标车辆可以为图像采集设备检测到的车辆。
图像采集设备采集到包含目标车辆的图像后,为了获得目标车辆更多的信息,如,车辆品牌信息、归属地等,图像采集设备可以识别出目标车辆的车牌信息,并将其识别出的车牌信息和对应的图像一起发送给图像分析系统进行分析。
通常情况下,图像采集设备采集的图像中可能会包含多辆车,但是需要分析的目标车辆只有一辆。因此,图像分析系统接收到图像采集设备发送的车牌信息和对应的图像后,可以根据车牌信息识别出图像中包含的目标车辆,进而对识别出的目标车辆进行分析。
上述方法中,图像分析系统在对目标车辆进行分析之前,需要首先对整个图像进行分析,识别出其中包含的目标车辆。而这将耗费较长的时间,导致图像分析效率较低。
发明内容
本申请实施例的目的在于提供一种图像分析方法及装置,以提高图像分析效率。具体技术方案如下:
第一方面,本申请实施例提供了一种图像分析方法,所述方法包括:
接收目标图像采集设备发送的目标信息;其中,所述目标信息至少包括:所述目标图像采集设备采集的第一图像、以及所述目标图像采集设备的标识信息;
根据预先保存的各图像采集设备的标识信息与各区域信息的对应关系,获取与所述目标图像采集设备的标识信息对应的目标区域信息;
在所述第一图像中,确定与所述目标区域信息对应的第二图像,并对所述第二图像进行分析。
可选地,所述目标信息还包括:所述第一图像中包括的目标车辆的第一车牌信息;所述对所述第二图像进行分析至少包括:
识别所述第二图像中的第二车牌信息;
所述方法还包括:
判断所述第一车牌信息与所述第二车牌信息是否相同;
如果是,对所述第二图像进行分析,确定所述第二图像的第一区域信息,并对应保存所述目标图像采集设备的标识信息、所述第二图像的第一区域信息、以及第一判断结果,其中,所述第一判断结果为所述第一车牌信息与所述第二车牌信息相同;
如果否,对所述第一图像进行分析,确定所述第一图像的第一区域信息,并对应保存所述目标图像采集设备的标识信息、所述第一图像的第一区域信息、以及第二判断结果,其中,所述第二判断结果为所述第一车牌信息与所述第二车牌信息不同。
可选地,所述确定所述第二图像的第一区域信息的步骤包括:
识别所述第二图像中车牌信息为所述第一车牌信息的目标车辆;
确定所述目标车辆在所述第二图像中所占区域的位置信息,并将该位置信息确定为所述第二图像的第一区域信息。
可选地,所述确定所述第一图像的第一区域信息的步骤包括:
识别所述第一图像中车牌信息为所述第一车牌信息的目标车辆;
确定所述目标车辆在所述第一图像中所占区域的位置信息,并将该位置信息确定为所述第一图像的第一区域信息。
可选地,所述方法还包括:
按照设定的时间间隔,判断在第一预设时间段内保存的所述第二判断结果的数量是否大于第一预设阈值;
如果是,根据第二预设时间段内保存的各第二判断结果对应的所述目标图像采集设备的标识信息、所述第一图像的第一区域信息,更新所述目标图像采集设备的标识信息对应的区域信息;
如果否,根据第三预设时间段内保存的各第一判断结果对应的所述目标图像采集设备的标识信息、所述第二图像的第一区域信息,更新所述目标图像采集设备的标识信息对应的区域信息。
可选地,所述根据第二预设时间段内保存的各第二判断结果对应的所述目标图像采集设备的标识信息、所述第一图像的第一区域信息,更新所述目标图像采集设备的标识信息对应的区域信息的步骤包括:
从所述第二预设时间段内保存的各第二判断结果中,获取所述目标图像采集设备的标识信息对应的多个第一图像的第一区域信息;
基于所述多个第一图像的第一区域信息的统计结果,或基于所述多个第一图像的第一区域信息的特征匹配结果,确定第二区域信息;
根据所确定的第二区域信息,更新所述目标图像采集设备的标识信息对应的区域信息。
可选地,所述根据第三预设时间段内保存的各第一判断结果对应的所述目标图像采集设备的标识信息、所述第二图像的第一区域信息,更新所述目标图像采集设备的标识信息对应的区域信息的步骤包括:
从所述第三预设时间段内保存的各第一判断结果中,获取所述目标图像采集设备的标识信息对应的多个第二图像的第一区域信息;
基于所述多个第二图像的第一区域信息的统计结果,或基于所述多个第二图像的第一区域信息的特征匹配结果,确定第二区域信息;
根据所确定的第二区域信息,更新所述目标图像采集设备的标识信息对应的区域信息。
可选地,预先保存各图像采集设备的标识信息与各区域信息的对应关系的步骤包括:
针对各图像采集设备,接收该图像采集设备发送的第三图像、所述第三图像中包括的目标车辆的第三车牌信息、以及该图像采集设备的标识信息;
识别所述第三图像中包括的车牌信息为所述第三车牌信息的目标车辆,并确定该目标车辆在所述第三图像中所占区域的位置信息,将该位置信息确定为所述第三图像的初始区域信息;
保存该图像采集设备的标识信息与该初始区域信息的对应关系;
当该图像采集设备对应的初始区域信息的数量大于第二预设阈值时,根据该图像采集设备的标识信息对应的各初始区域信息,确定并保存该图像采集设备对应的区域信息。
第二方面,本申请实施例提供了一种图像分析装置,所述装置包括:
第一接收模块,用于接收目标图像采集设备发送的目标信息;其中,所述目标信息至少包括:所述目标图像采集设备采集的第一图像、以及所述目标图像采集设备的标识信息;
执行模块,用于根据预先保存的各图像采集设备的标识信息与各区域信息的对应关系,获取与所述目标图像采集设备的标识信息对应的目标区域信息;
分析模块,用于在所述第一图像中,确定与所述目标区域信息对应的第二图像,并对所述第二图像进行分析。
可选地,所述目标信息还包括:所述第一图像中包括的目标车辆的第一车牌信息;所述分析模块,具体用于识别所述第二图像中的第二车牌信息;
所述装置还包括:
第一判断模块,用于判断所述第一车牌信息与所述第二车牌信息是否相同;
第一存储模块,用于当所述第一判断模块判断结果为是时,对所述第二图像进行分析,确定所述第二图像的第一区域信息,并对应保存所述目标图像采集设备的标识信息、所述第二图像的第一区域信息、以及第一判断结果,其中,所述第一判断结果为所述第一车牌信息与所述第二车牌信息相同;
第二存储模块,用于当所述第一判断模块判断结果为否时,对所述第一图像进行分析,确定所述第一图像的第一区域信息,并对应保存所述目标图像采集设备的标识信息、所述第一图像的第一区域信息、以及第二判断结果,其中,所述第二判断结果为所述第一车牌信息与所述第二车牌信息不同。
可选地,所述第一存储模块,包括:
第一识别子模块,用于识别所述第二图像中车牌信息为所述第一车牌信息的目标车辆;
第一确定子模块,用于确定所述目标车辆在所述第二图像中所占区域的位置信息,并将该位置信息确定为所述第二图像的第一区域信息。
可选地,所述第二存储模块,包括:
第二识别子模块,用于识别所述第一图像中车牌信息为所述第一车牌信息的目标车辆;
第二确定子模块,用于确定所述目标车辆在所述第一图像中所占区域的位置信息,并将该位置信息确定为所述第一图像的第一区域信息。
可选地,所述装置还包括:
第二判断模块,用于按照设定的时间间隔,判断在第一预设时间段内保存的所述第二判断结果的数量是否大于第一预设阈值;
第一更新模块,用于当所述第二判断模块判断结果为是时,根据第二预设时间段内保存的各第二判断结果对应的所述目标图像采集设备的标识信 息、所述第一图像的第一区域信息,更新所述目标图像采集设备的标识信息对应的区域信息;
第二更新模块,用于当所述第二判断模块判断结果为否时,根据第三预设时间段内保存的各第一判断结果对应的所述目标图像采集设备的标识信息、所述第二图像的第一区域信息,更新所述目标图像采集设备的标识信息对应的区域信息。
可选地,所述第一更新模块,包括:
第一获取子模块,用于从所述第二预设时间段内保存的各第二判断结果中,获取所述目标图像采集设备的标识信息对应的多个第一图像的第一区域信息;
第三确定子模块,用于基于所述多个第一图像的第一区域信息的统计结果,或基于所述多个第一图像的第一区域信息的特征匹配结果,确定第二区域信息;
第一更新子模块,用于根据所确定的第二区域信息,更新所述目标图像采集设备的标识信息对应的区域信息。
可选地,所述第二更新模块,包括:
第二获取子模块,用于从所述第三预设时间段内保存的各第一判断结果中,获取所述目标图像采集设备的标识信息对应的多个第二图像的第一区域信息;
第四确定子模块,用于基于所述多个第二图像的第一区域信息的统计结果,或基于所述多个第二图像的第一区域信息的特征匹配结果,确定第二区域信息;
第二更新子模块,用于根据所确定的第二区域信息,更新所述目标图像采集设备的标识信息对应的区域信息。
可选地,所述装置还包括:
第二接收模块,用于针对各图像采集设备,接收该图像采集设备发送的第三图像、所述第三图像中包括的目标车辆的第三车牌信息、以及该图像采 集设备的标识信息;
处理模块,用于识别所述第三图像中包括的车牌信息为所述第三车牌信息的目标车辆,并确定该目标车辆在所述第三图像中所占区域的位置信息,将该位置信息确定为所述第三图像的初始区域信息;
第三存储模块,用于保存该图像采集设备的标识信息与该初始区域信息的对应关系;
第四存储模块,用于当所述第三存储模块保存的该图像采集设备对应的初始区域信息的数量大于第二预设阈值时,根据该图像采集设备的标识信息对应的各初始区域信息,确定并保存该图像采集设备对应的区域信息。
第三方面,本申请提供了一种电子设备,包括:
处理器和存储器;
所述存储器存储可执行程序代码;
所述处理器通过读取所述存储器中存储的可执行程序代码来运行与所述可执行程序代码对应的程序,以用于在运行时执行本申请第一方面所述的一种图像分析方法。
本申请实施例提供了一种图像分析方法及装置,所述方法包括:接收目标图像采集设备发送的目标信息;其中,所述目标信息至少包括:所述目标图像采集设备采集的第一图像、以及所述目标图像采集设备的标识信息;根据预先保存的各图像采集设备的标识信息与各区域信息的对应关系,获取与所述目标图像采集设备的标识信息对应的目标区域信息;在所述第一图像中,确定与所述目标区域信息对应的第二图像,并对所述第二图像进行分析。
本申请实施例中,针对同一图像采集设备采集的图像,其中的目标车辆所处区域通常是相对固定的,因此,能够预先获取各图像采集设备的标识信息与各区域信息的对应关系,上述区域信息包括各图像采集设备采集的图像中目标车辆所处区域的位置信息。当接收到目标图像采集设备发送的图像后,可以根据目标图像采集设备对应的目标区域信息,确定与目标区域信息对应的第二图像,并只对第二图像进行分析,与第一图像相比,第二图像的尺寸 较小,因此,能够提高图像分析效率。
附图说明
为了更清楚地说明本申请实施例和现有技术的技术方案,下面对实施例和现有技术中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1为图像采集设备的图像采集区域与检测区域示意图;
图2为本申请实施例提供的一种图像分析方法的流程图;
图3(a)为本申请实施例的目标图像采集设备采集的第一图像示意图;
图3(b)为根据图3(a)所示的第一图像,以及目标区域信息确定的第二图像示意图;
图4为本申请实施例提供的一种图像分析方法的另一流程图;
图5为本申请实施例提供的一种图像分析方法的另一流程图;
图6为本申请实施例的根据各初始区域信息确定区域信息的示意图;
图7为本申请实施例提供的一种图像分析方法的另一流程图;
图8为本申请实施例提供的一种图像分析装置的结构示意图;
图9为本申请实施例提供的一种图像分析装置的另一结构示意图;
图10为本申请实施例提供的一种图像分析装置的另一结构示意图;
图11为本申请实施例提供的一种电子设备的结构示意图。
具体实施方式
为了提高图像分析效率,本申请实施例提供了一种图像分析方法及装置。
下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本申请一部分实施例,而 不是全部的实施例。基于本申请中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本申请保护的范围。
需要说明的是,在不冲突的情况下,本申请中的实施例及实施例中的特征可以相互组合。下面将参考附图并结合实施例来详细说明本申请。
在交通监控领域,在路口、卡口等地方安装的图像采集设备可以对其图像采集区域进行检测,当检测到有车辆通过时,抓拍得到包含目标车辆的监控图像。并且,通常情况下,针对每个图像采集设备,当其安装完成后,其所处位置、采集视角等参数就确定了,因此,其图像采集区域也会相对固定。
在本申请的一个实施例中,图像采集设备在进行车辆检测时,并不是在整个图像采集区域进行车辆检测,而是在一个比较固定、包含在其采集区域内的一个小的检测区域内进行车辆检测。只有当图像采集设备的检测区域内有车辆通过时,其才能检测到有车辆通过,当该车辆在检测区域以外的图像采集区域内时,图像采集设备并不能检测到该车辆。也就是说,检测区域才是图像采集设备的感兴趣区域,也是图像采集设备采集的图像中目标车辆所在的区域。
相应的,图像采集设备采集的图像中,也包含图像采集区域和检测区域。其中,图像采集区域可以为整个图像区域,检测区域即为对应图像采集设备实际检测区域的图像区域。
或者,在另一个实施例中,在对图像采集区域进行分析时,可以将检测区域作为车辆检测的首选区域,图像采集区域内的非检测区域作为车辆检测的次选区域。
如图1所示,图像采集设备的图像采集区域可以为区域110,检测区域为区域120。由图1可以看出,图像采集设备的图像采集区域110的范围通常比较大,而其检测区域,120的范围比较小。
在本申请实施例中,为了提高图像分析效率,针对各图像采集设备,可以预先根据各图像采集设备采集的图像,确定其区域信息,即确定其采集的图像中检测区域的位置信息,并对应保存各图像采集设备的标识信息与区域信息的对应关系。
具体地,可以由图像分析系统根据各图像采集设备采集的图像,确定其区域信息并保存;或者,也可以由另一自学习系统根据各图像采集设备采集的图像,确定其区域信息并保存。当由自学习系统保存各图像采集设备采集的标识信息与区域信息的对应关系时,图像分析系统可以与自学习系统建立连接关系。本申请实施例中,以图像分析系统根据各图像采集设备采集的图像,确定其区域信息并保存为例,来说明本实施例提供的图像分析方法。
并且,各图像采集设备的标识信息与区域信息的对应关系可以保存在图像分析系统本地,也可以保存在与图像分析系统建立连接关系的外部设备中。
保存各图像采集设备的标识信息与区域信息的对应关系时,针对各图像采集设备采集的图像,可以设定坐标系,如图1所示,可以将图像的一个顶点确定为原点O,并确定X轴、Y轴。保存各图像采集设备对应的区域信息时,可以相应保存检测区域的X方向的起始点、终点,以及Y方向的起始点、终点。其中,保存的X方向的起始点、终点,以及Y方向的起始点、终点,其单位可以为像素。
图像分析系统保存的各图像采集设备的标识信息与区域信息的对应关系可以如表1所示:
表1
标识信息 区域信息
01 X(120,143),Y(12,58)
02 X(112,232),Y(21,53)
03 X(23,83),Y(14,49)
本申请实施例提供了一种图像分析方法过程,如图2所示,该过程可以包括以下步骤:
S101,接收目标图像采集设备发送的目标信息;其中,所述目标信息至少包括:所述目标图像采集设备采集的第一图像、以及所述目标图像采集设备的标识信息。
本申请实施例提供的方法可以应用于图像分析系统。具体地,该图像分析系统可以运行于电子设备中,如台式计算机、便携式计算机、智能移动终端等。
在本申请实施例中,为了进行交通监控,可以在需要进行监控的路口、卡口等地方安装图像采集设备。其中,上述图像采集设备可以为球机、摄像机、抓拍机等,本申请实施例对此不进行限定。
并且,可以在图像采集设备与图像分析系统之间建立有线或无线连接,从而图像采集设备可以将其采集的图像发送给图像分析系统。例如,可以通过WIFI(Wireless Fidelity,无线保真)、NFC(Near Field Communication,近距离无线通讯技术)、蓝牙等无线连接方式在图像采集设备与图像分析系统之间建立连接,本申请实施例对此不进行限定。
需要说明的是,本申请实施例提供的方法可以对各个图像采集设备采集的图像进行分析,本实施例仅以任一图像采集设备为例,来说明本申请实施例提供的图像分析方法。并且,为了描述方便,可以将本申请实施例中针对的图像采集设备称为目标图像采集设备。
在本申请实施例中,目标图像采集设备可以检测其检测区域是否有车辆通过,当检测到时,可以采集包含其检测到的目标车辆的第一图像。并且,目标图像采集设备可以将其采集的第一图像,以及自身的标识信息发送给图像分析系统,以使图像分析系统获取更多的目标车辆的信息。
或者,如果在检测区域附近设置红外感应装置等车辆检测设备,目标采集设备也可以在接收到车辆检测设备的指示信号时采集检测区域的图像,例如,目标车辆驶过红外感应装置发出的红外射线时,确定目标车辆即将驶入检测区域,红外感应装置就向车辆检测设备发送指示信号,指示目标采集设备采集图像,而目标采集设备采集图像(如抓拍机抓拍图像)时,目标车辆也正位于驶过检测区域,目标采集设备就可以采集到目标车辆的第一图像。
又或者,在检测区域附近的路面上设置感应(如重量感应器),一旦车辆驶过该路面,感应装置就指示目标采集设备(如摄像机)采集视频或者图像,就可以采集到目标车辆的第一图像。
因此,在本申请实施例中,图像分析系统可以接收目标图像采集设备发送的目标信息,其中,该目标信息至少可以包括:目标图像采集设备采集的第一图像、以及目标图像采集设备的标识信息。例如,图像分析系统接收到的目标信息中包括的目标图像采集设备的标识信息可以为:02,图像分析系统接收的目标信息中包括的第一图像可以如图3(a)所示。
S102,根据预先保存的各图像采集设备的标识信息与各区域信息的对应关系,获取与所述目标图像采集设备的标识信息对应的目标区域信息。
在本申请实施例中,获取到目标图像采集设备发送的目标信息后,图像分析系统可以根据预先保存的各图像采集设备的标识信息与各区域信息的对应关系,获取与目标图像采集设备的标识信息对应的目标区域信息。具体地,图像分析系统可以在各图像采集设备的标识信息中,查找与目标图像采集设备的标识信息相同的标识信息,进而将识别到的标识信息对应的区域信息确定为目标区域信息。
例如,当图像分析系统获取到的各图像采集设备的标识信息与各区域信息的对应关系如表1所示,目标图像采集设备的标识信息为02时,其获取的与目标图像采集设备的标识信息对应的目标区域信息即为:X(112,232),Y(21,53)。
S103,在所述第一图像中,确定与所述目标区域信息对应的第二图像,并对所述第二图像进行分析。
在本申请实施例中,当图像分析系统获取到与目标图像采集设备匹配的目标区域信息后,其可以在第一图像中,确定与目标区域信息对应的第二图像,并对第二图像进行分析。
具体地,图像分析系统可以在第一图像中,确定目标区域信息对应的区域,并将包含该区域的图像确定为第二图像,进而可以对第二图像进行分析。
例如,当图像分析系统获取的第一图像如图3(a)所示,其识别出的目标区域信息为:X(112,232),Y(21,53)时,其可以确定第一图像中,X轴方向从112像素到232像素,Y轴方向从21像素到53像素的区域为对应目标区域信息的区域,如图3(a)中的区域310。进而,可以确定包含该区域的图像为 第二图像。图像采集设备确定的第二图像可以如图3(b)所示。
第二图像可以是包括目标车辆的图像,也可以是仅包括车牌的图像。
电子设备对第二图像进行分析时,例如可以识别第二图像中包括的目标车辆的车辆品牌、归属地等信息,本申请实施例对此不做限定。
本申请实施例中,针对同一图像采集设备采集的图像,其中的目标车辆所处区域通常是相对固定的,因此,能够预先获取各图像采集设备的标识信息与各区域信息的对应关系,上述区域信息包括各图像采集设备采集的图像中目标车辆所处区域的位置信息。当接收到目标图像采集设备发送的图像后,可以根据目标图像采集设备对应的目标区域信息,确定与目标区域信息对应的第二图像,并只对第二图像进行分析,与第一图像相比,第二图像的尺寸较小,因此,能够提高图像分析效率。
作为本申请实施例的一种实施方式,图像分析系统预先保存目标图像采集设备的标识信息与目标区域信息的对应关系时,如果目标区域信息是车辆的信息,且该目标区域信息对应的第二图像中包括的车辆体型较大,而同样大小的车辆在以后很长一段时间之内都未再出现过,这将导致保存的目标图像采集设备的标识信息对应的目标区域信息不够准确;或者,有些情况下,即使目标区域信息是车牌的信息,该目标区域信息对应得第二图像包含的仅为车牌不包含整个车辆,而车牌的大小较为固定,但是当目标图像采集设备因一些外在因素,如大风等,其检测区域改变时,预先保存的目标图像采集设备对应的目标区域信息可能就不再准确。
因此,在本申请实施例中,图像分析系统可以对目标图像采集设备对应的区域信息进行优化更新,以保证目标图像采集设备对应的区域信息的准确性,以进一步提高图像分析效率。
如图4所示,本申请实施例提供的图像分析方法,可以包括以下步骤:
S201,接收目标图像采集设备发送的目标信息;其中,所述目标信息至少包括:所述目标图像采集设备采集的第一图像、所述目标图像采集设备的标识信息、以及所述第一图像中包括的目标车辆的第一车牌信息。
在本申请实施例中,目标图像采集设备检测到其检测区域有目标车辆时,其可以识别目标车辆的车牌信息,并发送给图像分析系统,以使图像分析系统根据该车牌信息,识别对应的目标车辆,并对目标车辆进行分析。
其中,图像采集设备可以通过任一种图像识别方法,识别第一图像中包括的目标车辆的车牌信息,本申请实施例对此不进行赘述。
因此,在本申请实施例中,图像分析系统接收到的目标信息中可以包括:目标图像采集设备采集的第一图像、目标图像采集设备的标识信息、以及第一图像中包括的目标车辆的第一车牌信息。
S202,根据预先保存的各图像采集设备的标识信息与各区域信息的对应关系,获取与所述目标图像采集设备的标识信息对应的目标区域信息。
本步骤与图2所示实施例中步骤S102基本相同,在此不进行赘述。
S203,在所述第一图像中,确定与所述目标区域信息对应的第二图像,并识别所述第二图像中的第二车牌信息。
在本申请实施例中,图像分析系统对第二图像进行分析时,至少可以识别第二图像中的第二车牌信息,以根据该第二车牌信息,以及目标图像采集设备发送的目标车辆的第一车牌信息,确定目标区域信息是否准确。
S204,判断所述第一车牌信息与所述第二车牌信息是否相同;如果是,执行步骤S205,如果否,执行步骤S206。
图像分析系统识别出第二图像中包括的车辆的第二车牌信息后,其可以判断目标图像采集设备发送的第一车牌信息与该第二车牌信息是否相同。
S205,对所述第二图像进行分析,确定所述第二图像的第一区域信息,并对应保存所述目标图像采集设备的标识信息、所述第二图像的第一区域信息、以及第一判断结果,其中,所述第一判断结果为所述第一车牌信息与所述第二车牌信息相同。
当图像分析系统确定第一车牌信息与第二车牌信息相同时,表明预先保存的目标图像采集设备对应的区域信息是正确的。
但是,预先保存的目标图像采集设备对应的区域信息对应的检测区域, 可能比目标图像采集设备的实际检测区域大。因此,图像分析系统可以对第二图像进行分析,确定第二图像的第一区域信息。其中,第二图像的第一区域信息可以为第二图像中的车辆所在区域的位置信息,或者,也可以为第二图像中的车牌所在区域的位置信息。
例如,图像分析系统可以识别第二图像中车牌信息为第一车牌信息的目标车辆;并确定目标车辆在第二图像中所占区域的位置信息,将该位置信息确定为第二图像的第一区域信息。或者,图像分析系统可以识别第二图像中车牌信息为第一车牌信息的车牌;并确定该车牌在第二图像中所占区域的位置信息,将该位置信息确定为第二图像的第一区域信息。其中,图像分析系统保存的第一区域信息的格式可以与其预先保存的区域信息的格式相同。
确定第二图像的第一区域信息后,图像分析系统可以对应保存目标图像采集设备的标识信息、第二图像的第一区域信息、以及第一判断结果,其中,第一判断结果为第一车牌信息与第二车牌信息相同,以根据保存的信息,对目标图像采集设备对应的区域信息进行优化更新。
S206,对所述第一图像进行分析,确定所述第一图像的第一区域信息,并对应保存所述目标图像采集设备的标识信息、所述第一图像的第一区域信息、以及第二判断结果,其中,所述第二判断结果为所述第一车牌信息与所述第二车牌信息不同。
当图像分析系统确定第一车牌信息与第二车牌信息不同时,表明预先保存的目标图像采集设备对应的区域信息是不够准确的。这种情况下,图像分析系统可以对第一图像进行整体分析,确定第一图像的第一区域信息。其中,第一图像的第一区域信息可以为第一图像中的车辆所在区域的位置信息,或者,也可以为第一图像中的车牌所在区域的位置信息。
例如,图像分析系统可以识别第一图像中车牌信息为第一车牌信息的目标车辆;并确定目标车辆在第一图像中所占区域的位置信息,将该位置信息确定为第二图像的第一区域信息。或者,图像分析系统可以识别第一图像中车牌信息为第一车牌信息的车牌;并确定该车牌在第一图像中所占区域的位置信息,将该位置信息确定为第一图像的第一区域信息。
确定第一图像的第一区域信息后,图像分析系统可以对应保存目标图像采集设备的标识信息、第一图像的第一区域信息、以及第二判断结果,其中,第二判断结果为第一车牌信息与第二车牌信息不同,以根据保存的信息,对目标图像采集设备对应的区域信息进行优化更新。
本实施例中,图像分析系统可以获得目标图像采集设备发送的目标车辆的第一车牌信息,并识别第二图像中的第二车牌信息,当第一车牌信息与第二车牌信息相同时,确定并保存第二图像的第一区域信息;当第一车牌信息与第二车牌信息不同时,确定并保存第一图像的第一区域信息;从而,能够根据保存的信息,对目标图像采集设备对应的区域信息进行优化更新。
作为本申请实施例的一种实施方式,当图像分析系统预先确定并保存各图像采集设备与各区域信息的对应关系时,其还可以周期性对各图像采集设备对应的区域信息进行更新。同样的,当自学习系统预先确定并保存各图像采集设备与各区域信息的对应关系时,其也可以按照类似的方式,周期性对各图像采集设备对应的区域信息进行更新。
在图4所示实施例的基础上,本申请实施例提供的图像分析方法,如图5所示,还可以包括:
S207,按照设定的时间间隔,判断在第一预设时间段内保存的所述第二判断结果的数量是否大于第一预设阈值;如果是,执行步骤S208,如果否,执行步骤S209。
在本申请实施例中,图像分析系统可以按照设定的时间间隔,如2分钟、5分钟、10分钟等,判断在第一预设时间段内保存的第二判断结果的数量是否大于第一预设阈值。上述第一预设时间段例如可以为2小时、5小时、12小时等,第一预设阈值例如可以为100、200、300等。
S208,根据第二预设时间段内保存的各第二判断结果对应的所述目标图像采集设备的标识信息、所述第一图像的第一区域信息,更新所述目标图像采集设备的标识信息对应的区域信息。
当图像分析系统判断出在第一预设时间段内保存的第二判断结果的数量 大于第一预设阈值时,表明在第一预设时间段内从第二图像中识别不出第一车牌信息的次数较多,从而可以确定目标图像采集设备对应的区域信息可能准确性较低,或已变化,这种情况下,可以对保存的目标图像采集设备对应的区域信息进行更新。
具体地,图像分析系统可以从第二预设时间段内保存的各第二判断结果中,获取目标图像采集设备的标识信息对应的多个第一图像的第一区域信息,然后基于多个第一图像的第一区域信息的统计结果,或基于多个第一图像的第一区域信息的特征匹配结果,确定第二区域信息,最后根据所确定的第二区域信息,更新目标图像采集设备的标识信息对应的区域信息。其中,上述第二预设时间段与第一预设时间段可以相同或不同。
在一种实现方式中,图像分析系统基于多个第一图像的第一区域信息的统计结果,确定第二区域信息的过程例如可以为,可以首先确定各第一区域信息对应的第一区域,然后确定覆盖各第一区域的第二区域,并将第二区域的位置信息确定为第二区域信息。如图6所示,当多个第一区域信息对应的第一区域分别为区域610、620、以及630时,确定的第二区域信息对应的第二区域可以为区域640。
图像分析系统还可以在多个第一区域信息中,统计出现概率最高的第一区域信息,并将该第一区域信息确定为第二区域信息。或者,图像分析系统还可以在多个第一区域信息对应的第一区域中,确定重叠率最高的第二区域,并将第二区域的位置信息确定为第二区域信息。或者,图像分析系统可以确定各第一区域信息对应的第一区域的中心点,然后以该中心点为中心,扩展到预设车牌大小的第二区域,并将第二区域的位置信息确定为第二区域信息。
在一种实现方式中,图像分析系统基于多个第一图像的第一区域信息的特征匹配结果,确定第二区域信息的过程,可以为基于尺度不变特征变换(Scale Invariant Feature Transform,SIFT)算法对多个第一图像的第一区域信息进行特征匹配,将特征匹配结果作为第二区域信息。
S209,根据第三预设时间段内保存的各第一判断结果对应的所述目标图像采集设备的标识信息、所述第二图像的第一区域信息,更新所述目标图像采集设备的标识信息对应的区域信息。
当图像分析系统判断出在第一预设时间段内保存的第二判断结果的数量不大于第一预设阈值时,表明在第一预设时间段内从第二图像中识别不出第一车牌信息的次数较少,从而可以确定目标图像采集设备对应的区域信息是比较准确的。
这种情况下,图像分析系统可以对目标图像采集设备对应的区域信息进行进一步优化。例如,图像分析系统可以从第三预设时间段内保存的各第一判断结果中,获取目标图像采集设备的标识信息对应的多个第二图像的第一区域信息,然后基于多个第二图像的第一区域信息的统计结果,或基于多个第二图像的第一区域信息的特征匹配结果,确定第二区域信息,最后根据所确定的第二区域信息,更新目标图像采集设备的标识信息对应的区域信息。其中,上述第三预设时间段与第一预设时间段可以相同或不同。
在一种实现方式中,图像分析系统基于多个第二图像的第一区域信息的统计结果,确定第二区域信息的过程例如可以为,可以首先确定各第一区域信息对应的第一区域,然后确定覆盖各第一区域的第二区域,并将第二区域的位置信息确定为第二区域信息。
图像分析系统还可以在多个第一区域信息中,统计出现概率最高的第一区域信息,并将该第一区域信息确定为第二区域信息。或者,图像分析系统还可以在多个第一区域信息对应的第一区域中,确定重叠率最高的第二区域,并将第二区域的位置信息确定为第二区域信息。或者,图像分析系统可以确定各第一区域信息对应的第一区域的中心点,然后以该中心点为中心,扩展到预设车牌大小的第二区域,并将第二区域的位置信息确定为第二区域信息。
在一种实现方式中,图像分析系统可以基于SIFT算法对多个第二图像的第一区域信息进行特征匹配,将特征匹配结果作为第二区域信息。
本申请实施例中,图像分析系统可以根据保存的第一图像的第一区域信息或第二图像的第一区域信息,对保存的各图像采集设备对应的区域信息进行优化更新,从而保证各图像采集设备对应的区域信息的准确性,进而提高图像分析效率。
在本申请实施例中,图像分析系统可以预先保存各图像采集设备与各区域信息的对应关系。具体地,如图7所示,本申请实施例提供的图像分析方法,还可以包括以下步骤:
S301,针对各图像采集设备,接收该图像采集设备发送的第三图像、所述第三图像中包括的目标车辆的第三车牌信息、以及该图像采集设备的标识信息。
在本申请实施例中,图像分析系统可以针对各图像采集设备,接收该图像采集设备发送的第三图像、第三图像中包括的目标车辆的第三车牌信息、以及该图像采集设备的标识信息。
其中,第三车牌信息是图像采集设备从第三图像中识别到的。例如,图像采集设备可以根据其检测区域,在第三图像对应区域中检测目标车辆,并识别目标车辆的第三车牌信息。
S302,识别所述第三图像中包括的车牌信息为所述第三车牌信息的目标车辆,并确定该目标车辆在所述第三图像中所占区域的位置信息,将该位置信息确定为所述第三图像的初始区域信息。
图像分析系统接收到图像采集设备发送的第三图像、第三车牌信息后,其可以识别第三图像中包括的车牌信息为第三车牌信息的目标车辆。例如,图像分析系统可以采用任一种图像识别方法,在第三图像中识别车牌信息为第三车牌信息的车辆,并将识别到的车牌确定为目标车辆。
识别出第三图像中包括的目标车辆后,图像分析系统可以进一步地确定目标车辆在第三图像中所占区域的位置信息。如,图像分析系统可以在第三图像中建立坐标系,进而确定目标车辆在第三图像中所占区域在X方向的起始点、终点,以及Y方向的起始点、终点,并将确定的位置信息确定为第三图像的初始区域信息。
S303,保存该图像采集设备的标识信息与该初始区域信息的对应关系。
确定第三图像的初始区域信息后,图像分析系统可以保存图像采集设备的标识信息与该初始区域信息的对应关系。例如,图像分析系统可以将图像采集设备的标识信息与初始区域信息保存在本地,或者,可以保存在与自身 建立连接关系的外部设备中。
可选地,图像分析系统可以针对各图像采集设备,分组保存各图像采集设备的标识信息对应的初始区域信息。例如,图像分析系统保存的各图像采集设备的标识信息与初始区域信息的对应关系可以如表2所示:
表2
Figure PCTCN2017107872-appb-000001
S304,当该图像采集设备对应的初始区域信息的数量大于第二预设阈值时,根据该图像采集设备的标识信息对应的各初始区域信息,确定并保存该图像采集设备对应的区域信息。
当该图像采集设备对应的初始区域信息的数量大于第二预设阈值,如,5000、10000、20000等时,图像分析系统可以根据该图像采集设备的标识信息对应的各初始区域信息,确定并保存该图像采集设备对应的区域信息。具体地,图像分析系统可以基于该图像采集设备对应的多个初始区域信息的统计结果或特征匹配结果,确定该图像采集设备对应的区域信息。
在一种实现方式中,图像分析系统基于该图像采集设备对应的多个初始区域信息的统计结果确定该图像采集设备对应的区域信息的过程例如可以为,可以首先确定各初始区域信息对应的初始区域,然后确定覆盖各初始区 域的目标区域,并将目标区域的位置信息确定为该图像采集设备对应的区域信息。
根据表2所示的各图像采集设备的标识信息与初始区域信息的对应关系,确定的各图像采集设备的标识信息与区域信息的对应关系可以如表3所示:
表3
标识信息 区域信息
01 X(110,151),Y(12,53)
02 X(112,235),Y(21,55)
03 X(23,86),Y(14,49)
图像分析系统还可以在多个初始区域信息中,统计出现概率最高的初始区域信息,并将该初始区域信息确定为该图像采集设备对应的区域信息。或者,图像分析系统还可以在多个初始区域信息对应的初始区域中,确定重叠率最高的目标区域,并将目标区域的位置信息确定为该图像采集设备对应的区域信息。或者,图像分析系统可以确定各初始区域信息对应的初始区域的中心点,然后以该中心点为中心,扩展到预设车牌大小的目标区域,并将目标区域的位置信息确定为该图像采集设备对应的区域信息。
在一种实现方式中,图像分析系统可以通过SIFT算法获取该图像采集设备对应的多个初始区域信息的特征匹配结果,作为该图像采集设备对应的区域信息。
本实施例中,图像分析系统可以预先保存各图像采集设备与各区域信息的对应关系,进而,在接收到目标图像采集设备发送的第一图像时,可以确定与目标图像采集设备的区域信息对应的第二图像,并仅对第二图像进行分析,与第一图像相比,第二图像的尺寸较小,因此,能够提高图像分析效率。
相应于上面的方法实施例,本申请实施例还提供了相应的装置实施例。
图8为本申请实施例提供的一种图像分析装置,所述装置包括:
第一接收模块810,用于接收目标图像采集设备发送的目标信息;其中,所述目标信息至少包括:所述目标图像采集设备采集的第一图像、以及所述目标图像采集设备的标识信息;
执行模块820,用于根据预先保存的各图像采集设备的标识信息与各区域信息的对应关系,获取与所述目标图像采集设备的标识信息对应的目标区域信息;
分析模块830,用于在所述第一图像中,确定与所述目标区域信息对应的第二图像,并对所述第二图像进行分析。
本申请实施例中,针对同一图像采集设备采集的图像,其中的目标车辆所处区域通常是相对固定的,因此,能够预先获取各图像采集设备的标识信息与各区域信息的对应关系,上述区域信息可以各图像采集设备采集的图像中目标车辆所处区域的位置信息。当接收到目标图像采集设备发送的图像后,可以根据目标图像采集设备对应的目标区域信息,确定与目标区域信息对应的第二图像,并只对第二图像进行分析,与第一图像相比,第二图像的尺寸较小,因此,能够提高图像分析效率。
作为本申请实施例的一种实施方式,所述目标信息还包括:所述第一图像中包括的目标车辆的第一车牌信息;所述分析模块,具体用于识别所述第二图像中的第二车牌信息;如图9所示,所述装置还包括:
第一判断模块840,用于判断所述第一车牌信息与所述第二车牌信息是否相同;
第一存储模块850,用于当所述第一判断模块840判断结果为是时,对所述第二图像进行分析,确定所述第二图像的第一区域信息,并对应保存所述目标图像采集设备的标识信息、所述第二图像的第一区域信息、以及第一判断结果,其中,所述第一判断结果为所述第一车牌信息与所述第二车牌信息相同;
第二存储模块860,用于当所述第一判断模块840判断结果为否时,对所述第一图像进行分析,确定所述第一图像的第一区域信息,并对应保存所述目标图像采集设备的标识信息、所述第一图像的第一区域信息、以及第二判断结果,其中,所述第二判断结果为所述第一车牌信息与所述第二车牌信息不同。
作为本申请实施例的一种实施方式,所述第一存储模块850,包括:
第一识别子模块(图中未示出),用于识别所述第二图像中车牌信息为所述第一车牌信息的目标车辆;
第一确定子模块(图中未示出),用于确定所述目标车辆在所述第二图像中所占区域的位置信息,并将该位置信息确定为所述第二图像的第一区域信息。
作为本申请实施例的一种实施方式,所述第二存储模块860,包括:
第二识别子模块(图中未示出),用于识别所述第一图像中车牌信息为所述第一车牌信息的目标车辆;
第二确定子模块(图中未示出),用于确定所述目标车辆在所述第一图像中所占区域的位置信息,并将该位置信息确定为所述第一图像的第一区域信息。
作为本申请实施例的一种实施方式,如图9所示,所述装置还包括:
第二判断模块870,用于按照设定的时间间隔,判断在第一预设时间段内保存的所述第二判断结果的数量是否大于第一预设阈值;
第一更新模块880,用于当所述第二判断模块870判断结果为是时,根据第二预设时间段内保存的各第二判断结果对应的所述目标图像采集设备的标识信息、所述第一图像的第一区域信息,更新所述目标图像采集设备的标识信息对应的区域信息;
第二更新模块890,用于当所述第二判断模块870判断结果为否时,根据第三预设时间段内保存的各第一判断结果对应的所述目标图像采集设备的标识信息、所述第二图像的第一区域信息,更新所述目标图像采集设备的标识 信息对应的区域信息。
作为本申请实施例的一种实施方式,所述第一更新模块880,包括:
第一获取子模块(图中未示出),用于从所述第二预设时间段内保存的各第二判断结果中,获取所述目标图像采集设备的标识信息对应的多个第一图像的第一区域信息;
第三确定子模块(图中未示出),用于基于所述多个第一图像的第一区域信息的统计结果,或基于所述多个第一图像的第一区域信息的特征匹配结果,确定第二区域信息;
第一更新子模块(图中未示出),用于根据所确定的第二区域信息,更新所述目标图像采集设备的标识信息对应的区域信息。
作为本申请实施例的一种实施方式,所述第二更新模块890,包括:
第二获取子模块(图中未示出),用于从所述第三预设时间段内保存的各第一判断结果中,获取所述目标图像采集设备的标识信息对应的多个第二图像的第一区域信息;
第四确定子模块(图中未示出),用于基于所述多个第二图像的第一区域信息的统计结果,或基于所述多个第二图像的第一区域信息的特征匹配结果,确定第二区域信息;
第二更新子模块(图中未示出),用于根据所确定的第二区域信息,更新所述目标图像采集设备的标识信息对应的区域信息。
作为本申请实施例的一种实施方式,如图10所示,所述装置还包括:
第二接收模块1010,用于针对各图像采集设备,接收该图像采集设备发送的第三图像、所述第三图像中包括的目标车辆的第三车牌信息、以及该图像采集设备的标识信息;
处理模块1020,用于识别所述第三图像中包括的车牌信息为所述第三车牌信息的目标车辆,并确定该目标车辆在所述第三图像中所占区域的位置信息,将该位置信息确定为所述第三图像的初始区域信息;
第三存储模块1030,用于保存该图像采集设备的标识信息与该初始区域信息的对应关系;
第四存储模块1040,用于当所述第三存储模块1030保存的该图像采集设备对应的初始区域信息的数量大于第二预设阈值时,根据该图像采集设备的标识信息对应的各初始区域信息,确定并保存该图像采集设备对应的区域信息。
本申请实施例中,能够仅对包含第一图像特定区域的第二图像进行分析,与第一图像相比,第二图像的尺寸较小,因此,能够提高图像分析效率。
本申请实施例还提供了一种电子设备,可以包括:处理器和存储器;
所述存储器存储可执行程序代码;
所述处理器通过读取所述存储器中存储的可执行程序代码来运行与所述可执行程序代码对应的程序,以用于执行上述图像分析方法。
相应地,如图11所示,本申请实施例还提供了一种电子设备,可以包括:
处理器1110、存储器1120、通信接口1130和总线1140;
所述处理器1110、所述存储器1120和所述通信接口1130通过所述总线1140连接并完成相互间的通信;
所述存储器1120存储可执行程序代码;
所述处理器1110通过读取所述存储器1120中存储的可执行程序代码来运行与所述可执行程序代码对应的程序,以用于在运行时执行本申请实施例所述的一种图像分析方法,其中,所述图像分析方法包括:
接收目标图像采集设备发送的目标信息;其中,所述目标信息至少包括:所述目标图像采集设备采集的第一图像、以及所述目标图像采集设备的标识信息;
根据预先保存的各图像采集设备的标识信息与各区域信息的对应关系,获取与所述目标图像采集设备的标识信息对应的目标区域信息;
在所述第一图像中,确定与所述目标区域信息对应的第二图像,并对所述第二图像进行分析。
本申请实施例中,针对同一图像采集设备采集的图像,其中的目标车辆所处区域通常是相对固定的,因此,能够预先获取各图像采集设备的标识信息与各区域信息的对应关系,上述区域信息包括各图像采集设备采集的图像中目标车辆所处区域的位置信息。当接收到目标图像采集设备发送的图像后,可以根据目标图像采集设备对应的目标区域信息,确定与目标区域信息对应的第二图像,并只对第二图像进行分析,与第一图像相比,第二图像的尺寸较小,因此,能够提高图像分析效率。
相应地,本申请实施例还提供了一种存储介质,其中,该存储介质用于存储可执行程序代码,所述可执行程序代码用于在运行时执行本申请实施例所述的一种图像分析方法,其中,所述图像分析方法包括:
接收目标图像采集设备发送的目标信息;其中,所述目标信息至少包括:所述目标图像采集设备采集的第一图像、以及所述目标图像采集设备的标识信息;
根据预先保存的各图像采集设备的标识信息与各区域信息的对应关系,获取与所述目标图像采集设备的标识信息对应的目标区域信息;
在所述第一图像中,确定与所述目标区域信息对应的第二图像,并对所述第二图像进行分析。
本申请实施例中,针对同一图像采集设备采集的图像,其中的目标车辆所处区域通常是相对固定的,因此,能够预先获取各图像采集设备的标识信息与各区域信息的对应关系,上述区域信息包括各图像采集设备采集的图像中目标车辆所处区域的位置信息。当接收到目标图像采集设备发送的图像后,可以根据目标图像采集设备对应的目标区域信息,确定与目标区域信息对应的第二图像,并只对第二图像进行分析,与第一图像相比,第二图像的尺寸较小,因此,能够提高图像分析效率。
相应地,本申请实施例还提供了一种应用程序,其中,该应用程序用于在运行时执行本申请实施例所述的一种图像分析方法,其中,所述图像分析 方法包括:
接收目标图像采集设备发送的目标信息;其中,所述目标信息至少包括:所述目标图像采集设备采集的第一图像、以及所述目标图像采集设备的标识信息;
根据预先保存的各图像采集设备的标识信息与各区域信息的对应关系,获取与所述目标图像采集设备的标识信息对应的目标区域信息;
在所述第一图像中,确定与所述目标区域信息对应的第二图像,并对所述第二图像进行分析。
本申请实施例中,针对同一图像采集设备采集的图像,其中的目标车辆所处区域通常是相对固定的,因此,能够预先获取各图像采集设备的标识信息与各区域信息的对应关系,上述区域信息包括各图像采集设备采集的图像中目标车辆所处区域的位置信息。当接收到目标图像采集设备发送的图像后,可以根据目标图像采集设备对应的目标区域信息,确定与目标区域信息对应的第二图像,并只对第二图像进行分析,与第一图像相比,第二图像的尺寸较小,因此,能够提高图像分析效率。
对于装置/电子设备/存储介质/应用程序实施例而言,由于其基本相似于方法实施例,所以描述的比较简单,相关之处参见方法实施例的部分说明即可。
需要说明的是,在本文中,诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来,而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
本说明书中的各个实施例均采用相关的方式描述,各个实施例之间相同 相似的部分互相参见即可,每个实施例重点说明的都是与其他实施例的不同之处。尤其,对于装置实施例而言,由于其基本相似于方法实施例,所以描述的比较简单,相关之处参见方法实施例的部分说明即可。
本领域普通技术人员可以理解实现上述方法实施方式中的全部或部分步骤是可以通过程序来指令相关的硬件来完成,所述的程序可以存储于计算机可读取存储介质中,这里所称得的存储介质,如:ROM/RAM、磁碟、光盘等。
以上所述仅为本申请的较佳实施例而已,并非用于限定本申请的保护范围。凡在本申请的精神和原则之内所作的任何修改、等同替换、改进等,均包含在本申请的保护范围内。

Claims (17)

  1. 一种图像分析方法,其特征在于,所述方法包括:
    接收目标图像采集设备发送的目标信息;其中,所述目标信息至少包括:所述目标图像采集设备采集的第一图像、以及所述目标图像采集设备的标识信息;
    根据预先保存的各图像采集设备的标识信息与各区域信息的对应关系,获取与所述目标图像采集设备的标识信息对应的目标区域信息;
    在所述第一图像中,确定与所述目标区域信息对应的第二图像,并对所述第二图像进行分析。
  2. 根据权利要求1所述的方法,其特征在于,所述目标信息还包括:所述第一图像中包括的目标车辆的第一车牌信息;所述对所述第二图像进行分析至少包括:
    识别所述第二图像中的第二车牌信息;
    所述方法还包括:
    判断所述第一车牌信息与所述第二车牌信息是否相同;
    如果是,对所述第二图像进行分析,确定所述第二图像的第一区域信息,并对应保存所述目标图像采集设备的标识信息、所述第二图像的第一区域信息、以及第一判断结果,其中,所述第一判断结果为所述第一车牌信息与所述第二车牌信息相同;
    如果否,对所述第一图像进行分析,确定所述第一图像的第一区域信息,并对应保存所述目标图像采集设备的标识信息、所述第一图像的第一区域信息、以及第二判断结果,其中,所述第二判断结果为所述第一车牌信息与所述第二车牌信息不同。
  3. 根据权利要求2所述的方法,其特征在于,所述确定所述第二图像的第一区域信息的步骤包括:
    识别所述第二图像中车牌信息为所述第一车牌信息的目标车辆;
    确定所述目标车辆在所述第二图像中所占区域的位置信息,并将该位置信息确定为所述第二图像的第一区域信息。
  4. 根据权利要求2所述的方法,其特征在于,所述确定所述第一图像的第一区域信息的步骤包括:
    识别所述第一图像中车牌信息为所述第一车牌信息的目标车辆;
    确定所述目标车辆在所述第一图像中所占区域的位置信息,并将该位置信息确定为所述第一图像的第一区域信息。
  5. 根据权利要求2所述的方法,其特征在于,所述方法还包括:
    按照设定的时间间隔,判断在第一预设时间段内保存的所述第二判断结果的数量是否大于第一预设阈值;
    如果是,根据第二预设时间段内保存的各第二判断结果对应的所述目标图像采集设备的标识信息、所述第一图像的第一区域信息,更新所述目标图像采集设备的标识信息对应的区域信息;
    如果否,根据第三预设时间段内保存的各第一判断结果对应的所述目标图像采集设备的标识信息、所述第二图像的第一区域信息,更新所述目标图像采集设备的标识信息对应的区域信息。
  6. 根据权利要求5所述的方法,其特征在于,所述根据第二预设时间段内保存的各第二判断结果对应的所述目标图像采集设备的标识信息、所述第一图像的第一区域信息,更新所述目标图像采集设备的标识信息对应的区域信息的步骤包括:
    从所述第二预设时间段内保存的各第二判断结果中,获取所述目标图像采集设备的标识信息对应的多个第一图像的第一区域信息;
    基于所述多个第一图像的第一区域信息的统计结果,或基于所述多个第一图像的第一区域信息的特征匹配结果,确定第二区域信息;
    根据所确定的第二区域信息,更新所述目标图像采集设备的标识信息对应的区域信息。
  7. 根据权利要求5所述的方法,其特征在于,所述根据第三预设时间段内保存的各第一判断结果对应的所述目标图像采集设备的标识信息、所述第二图像的第一区域信息,更新所述目标图像采集设备的标识信息对应的区域信息的步骤包括:
    从所述第三预设时间段内保存的各第一判断结果中,获取所述目标图像采集设备的标识信息对应的多个第二图像的第一区域信息;
    基于所述多个第二图像的第一区域信息的统计结果,或基于所述多个第二图像的第一区域信息的特征匹配结果,确定第二区域信息;
    根据所确定的第二区域信息,更新所述目标图像采集设备的标识信息对应的区域信息。
  8. 根据权利要求1-7任一项所述的方法,其特征在于,预先保存各图像采集设备的标识信息与各区域信息的对应关系的步骤包括:
    针对各图像采集设备,接收该图像采集设备发送的第三图像、所述第三图像中包括的目标车辆的第三车牌信息、以及该图像采集设备的标识信息;
    识别所述第三图像中包括的车牌信息为所述第三车牌信息的目标车辆,并确定该目标车辆在所述第三图像中所占区域的位置信息,将该位置信息确定为所述第三图像的初始区域信息;
    保存该图像采集设备的标识信息与该初始区域信息的对应关系;
    当该图像采集设备对应的初始区域信息的数量大于第二预设阈值时,根据该图像采集设备的标识信息对应的各初始区域信息,确定并保存该图像采集设备对应的区域信息。
  9. 一种图像分析装置,其特征在于,所述装置包括:
    第一接收模块,用于接收目标图像采集设备发送的目标信息;其中,所述目标信息至少包括:所述目标图像采集设备采集的第一图像、以及所述目标图像采集设备的标识信息;
    执行模块,用于根据预先保存的各图像采集设备的标识信息与各区域信息的对应关系,获取与所述目标图像采集设备的标识信息对应的目标区域信 息;
    分析模块,用于在所述第一图像中,确定与所述目标区域信息对应的第二图像,并对所述第二图像进行分析。
  10. 根据权利要求9所述的装置,其特征在于,所述目标信息还包括:所述第一图像中包括的目标车辆的第一车牌信息;所述分析模块,具体用于识别所述第二图像中的第二车牌信息;
    所述装置还包括:
    第一判断模块,用于判断所述第一车牌信息与所述第二车牌信息是否相同;
    第一存储模块,用于当所述第一判断模块判断结果为是时,对所述第二图像进行分析,确定所述第二图像的第一区域信息,并对应保存所述目标图像采集设备的标识信息、所述第二图像的第一区域信息、以及第一判断结果,其中,所述第一判断结果为所述第一车牌信息与所述第二车牌信息相同;
    第二存储模块,用于当所述第一判断模块判断结果为否时,对所述第一图像进行分析,确定所述第一图像的第一区域信息,并对应保存所述目标图像采集设备的标识信息、所述第一图像的第一区域信息、以及第二判断结果,其中,所述第二判断结果为所述第一车牌信息与所述第二车牌信息不同。
  11. 根据权利要求10所述的装置,其特征在于,所述第一存储模块,包括:
    第一识别子模块,用于识别所述第二图像中车牌信息为所述第一车牌信息的目标车辆;
    第一确定子模块,用于确定所述目标车辆在所述第二图像中所占区域的位置信息,并将该位置信息确定为所述第二图像的第一区域信息。
  12. 根据权利要求10所述的装置,其特征在于,所述第二存储模块,包括:
    第二识别子模块,用于识别所述第一图像中车牌信息为所述第一车牌信息的目标车辆;
    第二确定子模块,用于确定所述目标车辆在所述第一图像中所占区域的位置信息,并将该位置信息确定为所述第一图像的第一区域信息。
  13. 根据权利要求10所述的装置,其特征在于,所述装置还包括:
    第二判断模块,用于按照设定的时间间隔,判断在第一预设时间段内保存的所述第二判断结果的数量是否大于第一预设阈值;
    第一更新模块,用于当所述第二判断模块判断结果为是时,根据第二预设时间段内保存的各第二判断结果对应的所述目标图像采集设备的标识信息、所述第一图像的第一区域信息,更新所述目标图像采集设备的标识信息对应的区域信息;
    第二更新模块,用于当所述第二判断模块判断结果为否时,根据第三预设时间段内保存的各第一判断结果对应的所述目标图像采集设备的标识信息、所述第二图像的第一区域信息,更新所述目标图像采集设备的标识信息对应的区域信息。
  14. 根据权利要求13所述的装置,其特征在于,所述第一更新模块,包括:
    第一获取子模块,用于从所述第二预设时间段内保存的各第二判断结果中,获取所述目标图像采集设备的标识信息对应的多个第一图像的第一区域信息;
    第三确定子模块,用于基于所述多个第一图像的第一区域信息的统计结果,或基于所述多个第一图像的第一区域信息的特征匹配结果,确定第二区域信息;
    第一更新子模块,用于根据所确定的第二区域信息,更新所述目标图像采集设备的标识信息对应的区域信息。
  15. 根据权利要求13所述的装置,其特征在于,所述第二更新模块,包括:
    第二获取子模块,用于从所述第三预设时间段内保存的各第一判断结果中,获取所述目标图像采集设备的标识信息对应的多个第二图像的第一区域 信息;
    第四确定子模块,用于基于所述多个第二图像的第一区域信息的统计结果,或基于所述多个第二图像的第一区域信息的特征匹配结果,确定第二区域信息;
    第二更新子模块,用于根据所确定的第二区域信息,更新所述目标图像采集设备的标识信息对应的区域信息。
  16. 根据权利要求9-15任一项所述的装置,其特征在于,所述装置还包括:
    第二接收模块,用于针对各图像采集设备,接收该图像采集设备发送的第三图像、所述第三图像中包括的目标车辆的第三车牌信息、以及该图像采集设备的标识信息;
    处理模块,用于识别所述第三图像中包括的车牌信息为所述第三车牌信息的目标车辆,并确定该目标车辆在所述第三图像中所占区域的位置信息,将该位置信息确定为所述第三图像的初始区域信息;
    第三存储模块,用于保存该图像采集设备的标识信息与该初始区域信息的对应关系;
    第四存储模块,用于当所述第三存储模块保存的该图像采集设备对应的初始区域信息的数量大于第二预设阈值时,根据该图像采集设备的标识信息对应的各初始区域信息,确定并保存该图像采集设备对应的区域信息。
  17. 一种电子设备,其特征在于,包括:
    处理器和存储器;
    所述存储器存储可执行程序代码;
    所述处理器通过读取所述存储器中存储的可执行程序代码来运行与所述可执行程序代码对应的程序,以用于执行如权利要求1-8任一项所述的一种图像分析方法。
PCT/CN2017/107872 2016-11-25 2017-10-26 一种图像分析方法及装置 WO2018095194A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US16/463,951 US11048950B2 (en) 2016-11-25 2017-10-26 Method and device for processing images of vehicles
EP17873492.7A EP3547253B1 (en) 2016-11-25 2017-10-26 Image analysis method and device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201611056740.2A CN108109132B (zh) 2016-11-25 2016-11-25 一种图像分析方法及装置
CN201611056740.2 2016-11-25

Publications (1)

Publication Number Publication Date
WO2018095194A1 true WO2018095194A1 (zh) 2018-05-31

Family

ID=62194751

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/107872 WO2018095194A1 (zh) 2016-11-25 2017-10-26 一种图像分析方法及装置

Country Status (4)

Country Link
US (1) US11048950B2 (zh)
EP (1) EP3547253B1 (zh)
CN (1) CN108109132B (zh)
WO (1) WO2018095194A1 (zh)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
SG11201908985YA (en) * 2017-03-30 2019-11-28 Mitsubishi Heavy Industries Machinery Systems Ltd Vehicle management system, vehicle-mounted device, vehicle management method, and program
WO2020062088A1 (zh) * 2018-09-28 2020-04-02 安徽继远软件有限公司 图像识别方法和设备、存储介质和处理器
CN111243281A (zh) * 2018-11-09 2020-06-05 杭州海康威视系统技术有限公司 一种道路多视频联合检测系统及检测方法
CN109815758A (zh) * 2019-01-17 2019-05-28 柳州康云互联科技有限公司 一种基于二维码的图像特征采集方法
CN111641805B (zh) * 2020-04-30 2021-10-01 武汉旷视金智科技有限公司 获取视频的方法、装置、终端设备和服务器
CN114706581A (zh) * 2022-04-06 2022-07-05 北京字节跳动网络技术有限公司 图像分析方法、装置、计算机设备以及存储介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103854026A (zh) * 2012-11-28 2014-06-11 联想(北京)有限公司 一种识别方法及电子设备
CN104966049A (zh) * 2015-06-01 2015-10-07 江苏大为科技股份有限公司 基于图像的货车检测方法
JP2016091193A (ja) * 2014-10-31 2016-05-23 キヤノン株式会社 画像処理装置、画像処理方法およびプログラム

Family Cites Families (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2236714C (en) * 1995-11-01 2005-09-27 Carl Kupersmit Vehicle speed monitoring system
EP1056064B1 (en) * 1999-05-28 2007-12-12 Nippon Telegraph and Telephone Corporation Apparatus and method for speed measurement of vehicles with an image processing system
US6587755B1 (en) * 2000-07-12 2003-07-01 International Business Machines Corporation Virtual signs for improving information communicated to the automotive driver
US6879263B2 (en) * 2000-11-15 2005-04-12 Federal Law Enforcement, Inc. LED warning light and communication system
US7124006B2 (en) * 2002-08-19 2006-10-17 Ron Yaacov Davidson Vehicle identification means detection and evasion system
US6888445B2 (en) * 2003-05-20 2005-05-03 Bradley L. Gotfried Vehicle identification system
US7843313B2 (en) * 2003-09-05 2010-11-30 Honeywell International Inc. Distributed stand-off verification and face recognition systems (FRS)
TW200937313A (en) * 2008-02-18 2009-09-01 Univ Nat Chiao Tung Method and system for increasing license plate detection efficiency in successively inputting image
US8229168B2 (en) * 2008-02-20 2012-07-24 International Business Machines Corporation Fast license plate verifier
US8698896B2 (en) 2012-08-06 2014-04-15 Cloudparc, Inc. Controlling vehicle use of parking spaces and parking violations within the parking spaces using multiple cameras
CN103810696B (zh) * 2012-11-15 2017-03-22 浙江大华技术股份有限公司 一种目标对象图像检测方法及装置
CN103402058B (zh) * 2013-08-22 2016-08-10 深圳市金立通信设备有限公司 一种拍摄图像的处理方法及装置
CN103699905B (zh) * 2013-12-27 2017-04-12 深圳市捷顺科技实业股份有限公司 一种车牌定位方法及装置
US10133530B2 (en) * 2014-05-19 2018-11-20 Allstate Insurance Company Electronic display systems connected to vehicles and vehicle-based systems
CN104239867B (zh) * 2014-09-17 2017-12-01 深圳市捷顺科技实业股份有限公司 车牌定位方法及系统
US20170046891A1 (en) * 2015-08-12 2017-02-16 Tyco Fire & Security Gmbh Systems and methods for location identification and tracking using a camera
CN106056915B (zh) * 2016-06-17 2019-07-23 京东方科技集团股份有限公司 车辆身份信息识别方法及系统、识别装置及采集装置
CN107103774A (zh) * 2017-06-19 2017-08-29 京东方科技集团股份有限公司 一种车辆监控方法和车辆监控装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103854026A (zh) * 2012-11-28 2014-06-11 联想(北京)有限公司 一种识别方法及电子设备
JP2016091193A (ja) * 2014-10-31 2016-05-23 キヤノン株式会社 画像処理装置、画像処理方法およびプログラム
CN104966049A (zh) * 2015-06-01 2015-10-07 江苏大为科技股份有限公司 基于图像的货车检测方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3547253A4

Also Published As

Publication number Publication date
EP3547253A1 (en) 2019-10-02
EP3547253B1 (en) 2024-01-10
US11048950B2 (en) 2021-06-29
CN108109132A (zh) 2018-06-01
CN108109132B (zh) 2021-02-05
US20200380277A1 (en) 2020-12-03
EP3547253A4 (en) 2019-12-11

Similar Documents

Publication Publication Date Title
WO2018095194A1 (zh) 一种图像分析方法及装置
US11257220B2 (en) Asset tracking systems
CN110473406B (zh) 搜索支援系统、搜索支援装置以及搜索支援方法
KR20180017606A (ko) 차량의 주차 위치 정보를 제공하기 위한 방법 및 그 전자 장치
WO2018000823A1 (zh) 一种导航方法、装置及系统
CN110568515B (zh) 基于红外阵列的人体存在检测方法、装置、存储介质
CN111291641A (zh) 图像识别方法及其装置、计算机可读介质和系统
JP4984640B2 (ja) 認識装置、地図データ作成装置および認識方法
CN111383246B (zh) 条幅检测方法、装置及设备
CN111666907A (zh) 一种视频中对象信息的识别方法、装置及服务器
KR101018033B1 (ko) 차선 이탈 경보 방법 및 시스템
CN111753587B (zh) 一种倒地检测方法及装置
CN108230680B (zh) 一种车辆行为信息获取方法、装置及终端
US11275963B2 (en) Image identification apparatus, image identification method, and non-transitory computer-readable storage medium for storing image identification program
JP2008077430A (ja) 移動体計数装置および移動体計数方法
KR101804471B1 (ko) 영상 분석 방법 및 장치
JP6772059B2 (ja) 電子制御装置、電子制御システムおよび電子制御方法
JP2008026999A (ja) 障害物検出システム、及び障害物検出方法
CN114463779A (zh) 吸烟识别方法、装置、设备及存储介质
WO2021140844A1 (ja) 人体検出装置および人体検出方法
JP2019046371A (ja) 交通標識を認識する車両制御装置及びその方法
KR101519966B1 (ko) 기준판 기반의 비전 인식 방법 및 시스템
JPWO2012160657A1 (ja) 物体接近検知装置、物体接近検知方法、及びプログラム
JP2017091381A (ja) セキュリティ装置、及び、セキュリティシステム
CN107680118B (zh) 一种影像辨识追踪方法

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: 17873492

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2017873492

Country of ref document: EP

Effective date: 20190625