WO2017161665A1 - 图像识别方法、装置、设备及非易失性计算机存储介质 - Google Patents
图像识别方法、装置、设备及非易失性计算机存储介质 Download PDFInfo
- Publication number
- WO2017161665A1 WO2017161665A1 PCT/CN2016/082970 CN2016082970W WO2017161665A1 WO 2017161665 A1 WO2017161665 A1 WO 2017161665A1 CN 2016082970 W CN2016082970 W CN 2016082970W WO 2017161665 A1 WO2017161665 A1 WO 2017161665A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- area
- images
- processing
- space
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000012545 processing Methods 0.000 claims abstract description 71
- 230000033001 locomotion Effects 0.000 claims description 37
- 238000003709 image segmentation Methods 0.000 claims description 22
- 230000008569 process Effects 0.000 claims description 13
- 238000003384 imaging method Methods 0.000 claims description 11
- 230000001133 acceleration Effects 0.000 description 5
- 238000001514 detection method Methods 0.000 description 5
- 238000012549 training Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 2
- 235000021443 coca cola Nutrition 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 229910044991 metal oxide Inorganic materials 0.000 description 1
- 150000004706 metal oxides Chemical class 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000004984 smart glass Substances 0.000 description 1
- 235000011888 snacks Nutrition 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/174—Segmentation; Edge detection involving the use of two or more images
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/61—Control of cameras or camera modules based on recognised objects
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/80—Camera processing pipelines; Components thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/68—Food, e.g. fruit or vegetables
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Definitions
- the present invention relates to image processing technologies, and in particular, to an image recognition method, apparatus, device, and nonvolatile computer storage medium.
- some handicrafts, drinks, snacks and other items can be placed in refrigerators, refrigerators, display cabinets, open shelves and other storage equipment for consumers to store, watch, purchase and so on.
- the staff can perform statistics on the site of the storage device to identify the basic situation of the internal space of the storage device, for example, the placement position of the article, the number of articles, etc., which complicates the operation and operates.
- the time is long and error-prone, resulting in a decrease in the efficiency and reliability of the identification.
- aspects of the present invention provide an image recognition method, apparatus, device, and non-volatile computer storage medium for improving efficiency and reliability of recognition.
- An aspect of the present invention provides an image recognition method, including:
- an implementation is further provided, the designated space comprising an interior space of the storage device.
- the performing image segmentation processing on the image to be identified to obtain at least one area image of the specified space including:
- image segmentation processing is performed on the image to be identified to obtain at least one area image of the designated space.
- acquiring the image to be identified in the specified space further includes:
- turning on and off of the imaging device is controlled to capture an image to be recognized of the designated space by using the imaging device.
- the acquiring the motion state of the door body of the storage device where the designated space is located including:
- the motion state of the door body is obtained.
- an image recognition apparatus comprising:
- An obtaining unit configured to acquire an image to be identified in a specified space
- a dividing unit configured to perform image segmentation processing on the image to be identified to obtain the Specifying at least one area image of the space
- a matching unit configured to perform image matching processing on each of the at least one area image to obtain a reference image corresponding to each of the area images
- an identifying unit configured to perform, according to image information of the reference image corresponding to each of the area images, an identification process on each of the area images to obtain item information of each of the area images.
- an implementation is further provided, the designated space comprising an interior space of the storage device.
- image segmentation processing is performed on the image to be identified to obtain at least one area image of the designated space.
- Image matching processing is performed in a reference image of the specified item collected in advance using each of the at least one area image to obtain a reference image corresponding to each of the area images.
- the device further comprising a control unit
- turning on and off of the imaging device is controlled to capture an image to be recognized of the designated space by using the imaging device.
- control unit is specifically configured to
- the motion state of the door body is obtained.
- an apparatus comprising:
- One or more processors are One or more processors;
- One or more programs the one or more programs being stored in the memory, when executed by the one or more processors:
- a nonvolatile computer storage medium storing one or more programs when the one or more programs are executed by a device causes The device:
- the embodiment of the present invention obtains an image to be identified in a specified space, and further performs image segmentation processing on the image to be identified to obtain at least one region image of the designated space, and the at least one region.
- Image matching processing is performed on each of the image images in the image to obtain a reference image corresponding to each of the region images, so that each of the region images can be obtained according to image information of the reference image corresponding to each of the region images.
- the identification process is performed to obtain the item information of the image of each area, without manual participation, the operation is simple, and the correct rate is high, thereby improving the efficiency and reliability of the recognition.
- the item information in the image to be recognized can be automatically recognized, which can effectively improve the efficiency of recognition, and can effectively improve the automation degree of the recognition.
- the change of the designated space is mainly caused by human operation, for example, the operator opens the door body of the storage device where the designated space is located, takes an item, etc., and therefore, according to The movement state of the door body of the storage device in which the designated space is located controls the opening and closing of the image capturing device to capture the image to be recognized in the designated space by using the image capturing device, thereby effectively reducing the power consumption of the image capturing device.
- FIG. 1 is a schematic flowchart of an image recognition method according to an embodiment of the present invention.
- FIG. 2 is a schematic structural diagram of an image recognition apparatus according to another embodiment of the present invention.
- FIG. 3 is a schematic structural diagram of an image recognition apparatus according to another embodiment of the present invention.
- the terminals involved in the embodiments of the present invention may include, but are not limited to, a mobile phone, a personal digital assistant (PDA), a wireless handheld device, a tablet computer, and a personal computer (Personal Computer, PC). ), MP3 player, MP4 player, wearable device (for example, smart glasses, smart watches, smart bracelets, etc.).
- PDA personal digital assistant
- PC Personal Computer
- FIG. 1 is a schematic flowchart of an image recognition method according to an embodiment of the present invention, as shown in FIG. 1 .
- the so-called designated space may refer to the internal space of the storage device, for example, the interior space of the display cabinet, the open shelf, the refrigerator, the refrigerator, and the like.
- the execution body of 101 to 104 may be an application located at a local terminal, or may be a plug-in or a software development kit (SDK) or the like provided in an application located in the local terminal, or It may also be a processing engine located in the network side server, or may also be a distributed system located on the network side, which is not specifically limited in this embodiment.
- SDK software development kit
- the application may be a local application (nativeApp) installed on the terminal, or may be a web application (webApp) of the browser on the terminal, which is not specifically limited in this embodiment.
- image segmentation processing is performed on the image to be recognized to obtain at least one region image of the specified space, and an image is performed on each region image in the at least one region image.
- Matching processing to obtain a reference image corresponding to each of the area images so that the image processing of each of the area images is performed according to image information of the reference image corresponding to each of the area images to obtain the
- the item information of each area image does not require manual participation, the operation is simple, and the correct rate is high, thereby improving the efficiency and reliability of the recognition.
- the so-called image refers to a certain image format
- the image data that is, the pixels of the image are stored in a certain manner
- the formed file may also be referred to as an image file.
- the image format of the image may include, but is not limited to, a bitmap (BMP) format, a Portable Network Graphic Format (PNG), and a Joint Photographic Experts Group (Joint Photographic Experts Group, The JPEG) format and the Exchangeable Image File Format (EXIF) are not particularly limited in this embodiment.
- BMP bitmap
- PNG Portable Network Graphic Format
- JPEG Joint Photographic Experts Group
- EXIF Exchangeable Image File Format
- the opening and closing of the camera device can be controlled according to the movement state of the door body of the storage device in which the designated space is located, so as to utilize the camera
- the device can capture the image to be recognized in the designated space, and can effectively reduce the power consumption of the imaging device. In addition, it is possible to effectively reduce the processing load required for the system to process the captured image to be recognized.
- the motion state of the door body of the storage device where the designated space is located may be further acquired, and then, according to the motion state, the motion state may be controlled.
- the imaging device is turned on and off to capture an image to be recognized of the designated space by using the imaging device.
- the image capturing device may employ an image sensor to acquire an image to be recognized in a specified space.
- the image sensor may be a Charge Coupled Device (CCD) sensor, or may be a Metal Oxide Semiconductor (CMOS) sensor, which is not particularly limited in this embodiment.
- CCD Charge Coupled Device
- CMOS Metal Oxide Semiconductor
- the camera device may be mounted on the door body of the storage device, or may be installed in the body of the storage device, and it is necessary to ensure that the articles inside the storage device cannot block the camera.
- the apparatus is not particularly limited in this embodiment.
- the sensor device disposed on the door body may be specifically used to obtain at least one of a motion parameter of the door body, for example, an acceleration, a rotational angular velocity, and a rotation angle.
- a motion parameter of the door body for example, an acceleration, a rotational angular velocity, and a rotation angle.
- the motion parameter the motion state of the door body is obtained.
- an acceleration sensor may be used to obtain an acceleration of the door body, or a gyroscope may be used to obtain a rotational angular velocity of the door body, or an acceleration sensor may also be utilized.
- the rotational angular velocity and the rotational angle of the door body are obtained, which is not particularly limited in this embodiment.
- the opening direction of the door body is a positive direction. If the rotation angular velocity is positive and greater than or equal to the preset opening threshold, it can be judged that the door body is open; after the door body is opened, the acceleration is less than or equal to a preset static threshold value, It is judged that the door body is stationary; if the rotation angular velocity is negative, and the rotation angle is less than or equal to the preset closing threshold value, it can be judged that the door body is closed.
- a start command can be sent to the camera device; when the door body is closed, a stop command can be sent to the camera device. After receiving the start command, the camera device starts capturing the image to be recognized in the designated space until a stop command is received.
- the camera device may start continuous shooting processing on the designated space at a preset shooting frequency to obtain a plurality of captured images of the designated space in the current shooting, and perform the captured images.
- Storage processing After receiving the stop command, from the stored captured images, one captured image is selected for policy processing to determine whether the captured image needs to be the image to be recognized for execution 101-104. If the image to be identified is required, a wireless communication module can be utilized, for example, a Wireless Fidelity (Wi-Fi) module, a Global System for Mobile Communications (GSM) module, and a universal
- Wi-Fi Wireless Fidelity
- GSM Global System for Mobile Communications
- GPRS General Packet Radio Service
- the adopted policy processing method is not the content that needs attention in this application, and details are not described herein again. Further, in order to save storage space, it is possible to further select no The other captured images selected are deleted.
- an image detector may be specifically used to perform image segmentation processing on the image to be identified to obtain at least one region image of the designated space.
- the training sample set can be pre-utilized for training to construct the item detector used.
- the training samples included in the training sample set may be labeled known samples, so that the known samples can be directly used for training to construct an article detector; or part of the labeled object has been marked. Knowing the sample and the other part is an unknown sample that has not been labeled, then you can use the known sample to train to build the initial item detector, and then use the original item detector to predict the unknown sample to obtain The detection result, and then the unknown sample can be labeled according to the detection result of the unknown sample to form a known sample, as a newly added known sample, and the newly added known sample and the original known sample are retrained.
- this embodiment does not make special Set.
- the item detector can obtain at least one area image of the designated space by using various methods, which is not specifically limited in this embodiment.
- One method may be to use a detection method based on a candidate image set to extract candidate images of a plurality of regions of the image to be identified, and to detect each candidate image, thereby obtaining whether each candidate image is an image of an article requiring detection of an article. If the candidate image is an item image in which an item needs to be detected, the candidate image is taken as one of the area images.
- Another method may be a method for directly detecting the positioning, and by directly locating the location of the item and the area where the item is located, detecting the position of the item in the image to be identified, thereby obtaining an area image corresponding to the item.
- an image matching process may be performed in a reference image of a predetermined item that is pre-acquired by using each of the at least one area image, A reference image corresponding to the image of each region is obtained.
- the image matching processing may be performed by using multiple matching methods in the prior art. For details, refer to related content in the prior art, and details are not described herein again. If the reference image is greater than or equal to the preset matching threshold, it can be determined that the region image is a real object to be recognized. If a reference image smaller than the matching threshold is matched, or if no reference image is matched, it may be determined that the region image is not the real object to be recognized.
- the image information of the reference image corresponding to each of the area images and the identification weight of the reference image may be specifically used for each of the The area image is subjected to recognition processing to obtain item information of each of the area images.
- each reference image may be specifically marked in advance to obtain image information of each reference image, for example, a basic attribute such as the name and origin of the item.
- image information of each reference image for example, a basic attribute such as the name and origin of the item.
- the recognition weights of the three reference images are equal.
- the image information of a reference image is the image of the Arctic Ocean, another reference image
- the image information is also the Arctic Ocean, and the image information of a reference image is Coca-Cola.
- the item information of the image of the region is obtained as the Arctic Ocean.
- the image to be identified in the specified space is acquired, and then the image to be recognized is subjected to image segmentation processing to obtain at least one region image of the specified space, and each region in the at least one region image.
- image segmentation processing to obtain at least one region image of the specified space, and each region in the at least one region image.
- the item information in the image to be recognized can be automatically recognized, which can effectively improve the efficiency of recognition, and can effectively improve the automation degree of the recognition.
- the change of the designated space is mainly caused by human operation, for example, the operator opens the door body of the storage device where the designated space is located, takes an item, etc., and therefore, according to The movement state of the door body of the storage device in which the designated space is located controls the opening and closing of the image capturing device to capture the image to be recognized in the designated space by using the image capturing device, thereby effectively reducing the power consumption of the image capturing device.
- FIG. 2 is a schematic structural diagram of an image recognition apparatus according to another embodiment of the present invention, as shown in FIG. 2 .
- the image recognition apparatus of the present embodiment may include an acquisition unit 21, a division unit 22, a matching unit 23, and an identification unit 24.
- the acquiring unit 21 is configured to acquire an image to be identified in the specified space
- the dividing unit 22 is configured to perform image segmentation processing on the image to be identified to obtain at least one region image of the specified space; Performing image matching processing on each of the at least one area image to obtain a reference image corresponding to each of the area images; and identifying unit 24 for using a reference image corresponding to each of the area images
- the image information is subjected to recognition processing for each of the area images to obtain item information of each of the area images.
- the so-called designated space may refer to the internal space of the storage device, for example, the interior space of the display cabinet, the open shelf, the refrigerator, the refrigerator, and the like.
- the image recognition apparatus may be an application located in a local terminal, or may be a plug-in or a software development kit (SDK) installed in an application located in the local terminal.
- SDK software development kit
- the unit may be a processing engine located in the network side server, or may be a distributed system located on the network side, which is not specifically limited in this embodiment.
- the application may be a local application (nativeApp) installed on the terminal, or may be a web application (webApp) of the browser on the terminal, which is not specifically limited in this embodiment.
- the dividing unit 22 Specifically, the image detector may be used to perform image segmentation processing on the image to be identified to obtain at least one region image of the designated space.
- the matching unit 23 may be specifically configured to use, in the reference image of the pre-acquired specified item, the image of each area in the at least one area image. Image matching processing to obtain a reference image corresponding to each of the area images.
- the identifying unit 24 may be specifically configured to: according to image information of a reference image corresponding to each area image and an identification weight of the reference image, The image of each area is subjected to recognition processing to obtain item information of the image of each of the areas.
- the image recognition apparatus provided in this embodiment may further include a control unit 31, which may be used to acquire the storage of the designated space. a motion state of the door body of the device; and controlling opening and closing of the image pickup device according to the motion state to capture an image to be recognized of the designated space by using the image pickup device.
- control unit 31 may be specifically configured to obtain a motion parameter of the door body by using a sensor device, and obtain a motion state of the door body according to the motion parameter.
- the image to be identified in the specified space is acquired by the acquiring unit, and then the image to be recognized is subjected to image segmentation processing by the segmentation unit to obtain at least one region image of the specified space, and the matching unit At least one area in each image Performing image matching processing on the image to obtain a reference image corresponding to each of the region images, so that the recognition unit can identify and process each of the region images according to the image information of the reference image corresponding to each of the region images.
- the recognition unit can identify and process each of the region images according to the image information of the reference image corresponding to each of the region images.
- the item information in the image to be recognized can be automatically recognized, which can effectively improve the efficiency of recognition, and can effectively improve the automation degree of the recognition.
- the change of the designated space is mainly caused by human operation, for example, the operator opens the door body of the storage device where the designated space is located, takes an item, etc., and therefore, according to The movement state of the door body of the storage device in which the designated space is located controls the opening and closing of the image capturing device to capture the image to be recognized in the designated space by using the image capturing device, thereby effectively reducing the power consumption of the image capturing device.
- the disclosed system, apparatus, and method may be implemented in other manners.
- the device embodiments described above are merely illustrative.
- the division of the unit is only a logical function division.
- there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
- the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be electrical, mechanical or otherwise. formula.
- the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
- each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
- the above integrated unit can be implemented in the form of hardware or in the form of hardware plus software functional units.
- the above-described integrated unit implemented in the form of a software functional unit can be stored in a computer readable storage medium.
- the above software functional unit is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor to perform the methods of the various embodiments of the present invention. Part of the steps.
- the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like, which can store program codes. .
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Studio Devices (AREA)
- Image Analysis (AREA)
Abstract
一种图像识别方法、装置、设备及非易失性计算机存储介质。通过获取指定空间的待识别图像(101),进而对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像(102),以及对所述至少一个区域图像中每个区域图像进行图像匹配处理,以获得所述每个区域图像所对应的参考图像(103),使得能够根据所述每个区域图像所对应的参考图像的图像信息,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息(104),无需人工参与,操作简单,而且正确率高,从而提高了识别的效率和可靠性。
Description
本申请要求了申请日为2016年03月23日,申请号为201610170233.5发明名称为“图像识别方法及装置”的中国专利申请的优先权。
本发明涉及图像处理技术,尤其涉及一种图像识别方法、装置、设备及非易失性计算机存储介质。
通常,可以将一些工艺品、饮品、零食等物品,放置在冰箱、冷藏柜、展示柜、开放式货架等储物设备中,以供消费者进行存储、观赏、选购等。
现有技术中,可以通工作人员在储物设备的现场进行统计,来识别该储物设备的内部空间的基本情况,例如,物品的摆放位置、物品数量等,这样会使得操作复杂,操作时间长,且容易出错,从而导致了识别的效率和可靠性的降低。
发明内容
本发明的多个方面提供一种图像识别方法、装置、设备及非易失性计算机存储介质,用以提高识别的效率和可靠性。
本发明的一方面,提供一种图像识别方法,包括:
获取指定空间的待识别图像;
对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像;
对所述至少一个区域图像中每个区域图像进行图像匹配处理,以获得所述每个区域图像所对应的参考图像;
根据所述每个区域图像所对应的参考图像的图像信息,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述指定空间包括储物设备的内部空间。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像,包括:
利用物品检测器,对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述对所述至少一个区域图像中每个区域图像进行图像匹配处理,以获得所述每个区域图像所对应的参考图像,包括:
利用所述至少一个区域图像中每个区域图像,在预先采集的指定物品的参考图像中进行图像匹配处理,以获得所述每个区域图像所对应的
参考图像。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述根据所述每个区域图像所对应的参考图像的图像信息,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息,包括:
根据所述每个区域图像所对应的参考图像的图像信息和该参考图像的识别权重,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述获取指定空间的待识别图像之前,还包括:
获取所述指定空间所在储物设备的门体的运动状态;以及
根据所述运动状态,控制摄像装置的开启和关闭,以利用所述摄像装置,拍摄所述指定空间的待识别图像。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述获取所述指定空间所在储物设备的门体的运动状态,包括:
利用传感器装置,获得所述门体的运动参数;以及
根据所述运动参数,获得所述门体的运动状态。
本发明的另一方面,提供一种图像识别装置,包括:
获取单元,用于获取指定空间的待识别图像;
分割单元,用于对所述待识别图像进行图像分割处理,以获得所述
指定空间的至少一个区域图像;
匹配单元,用于对所述至少一个区域图像中每个区域图像进行图像匹配处理,以获得所述每个区域图像所对应的参考图像;
识别单元,用于根据所述每个区域图像所对应的参考图像的图像信息,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述指定空间包括储物设备的内部空间。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述分割单元,具体用于
利用物品检测器,对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述匹配单元,具体用于
利用所述至少一个区域图像中每个区域图像,在预先采集的指定物品的参考图像中进行图像匹配处理,以获得所述每个区域图像所对应的参考图像。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述识别单元,具体用于
根据所述每个区域图像所对应的参考图像的图像信息和该参考图像的识别权重,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述装置还包括控制单元,用于
获取所述指定空间所在储物设备的门体的运动状态;以及
根据所述运动状态,控制摄像装置的开启和关闭,以利用所述摄像装置,拍摄所述指定空间的待识别图像。
如上所述的方面和任一可能的实现方式,进一步提供一种实现方式,所述控制单元,具体用于
利用传感器装置,获得所述门体的运动参数;以及
根据所述运动参数,获得所述门体的运动状态。
本发明的另一方面,提供一种设备,包括:
一个或者多个处理器;
存储器;
一个或者多个程序,所述一个或者多个程序存储在所述存储器中,当被所述一个或者多个处理器执行时:
获取指定空间的待识别图像;
对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像;
对所述至少一个区域图像中每个区域图像进行图像匹配处理,以获得所述每个区域图像所对应的参考图像;
根据所述每个区域图像所对应的参考图像的图像信息,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息。
本发明的另一方面,提供一种非易失性计算机存储介质,所述非易失性计算机存储介质存储有一个或者多个程序,当所述一个或者多个程序被一个设备执行时,使得所述设备:
获取指定空间的待识别图像;
对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像;
对所述至少一个区域图像中每个区域图像进行图像匹配处理,以获得所述每个区域图像所对应的参考图像;
根据所述每个区域图像所对应的参考图像的图像信息,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息。
由上述技术方案可知,本发明实施例通过获取指定空间的待识别图像,进而对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像,以及对所述至少一个区域图像中每个区域图像进行图像匹配处理,以获得所述每个区域图像所对应的参考图像,使得能够根据所述每个区域图像所对应的参考图像的图像信息,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息,无需人工参与,操作简单,而且正确率高,从而提高了识别的效率和可靠性。
另外,采用本发明所提供的技术方案,一旦获取指定空间的待识别图像,即能够自动识别出该待识别图像中的物品信息,能够有效提高识别的效率,而且能够有效提高识别的自动化程度。
另外,采用本发明所提供的技术方案,由于指定空间的变化主要是人为操作所导致,例如,操作者打开指定空间所在储物设备的门体,将某个物品拿走等,因此,可以根据所述指定空间所在储物设备的门体的运动状态,控制摄像装置的开启和关闭,以利用所述摄像装置,拍摄所述指定空间的待识别图像,能够有效降低摄像装置的功率消耗。
为了更清楚地说明本发明实施例中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作一简单地介绍,显而易见地,下面描述中的附图是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动性的前提下,还可以根据这些附图获得其他的附图。
图1为本发明一实施例提供的图像识别方法的流程示意图;
图2为本发明另一实施例提供的图像识别装置的结构示意图;
图3为本发明另一实施例提供的图像识别装置的结构示意图。
为使本发明实施例的目的、技术方案和优点更加清楚,下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整
地描述,显然,所描述的实施例是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的全部其他实施例,都属于本发明保护的范围。
需要说明的是,本发明实施例中所涉及的终端可以包括但不限于手机、个人数字助理(Personal Digital Assistant,PDA)、无线手持设备、平板电脑(Tablet Computer)、个人电脑(Personal Computer,PC)、MP3播放器、MP4播放器、可穿戴设备(例如,智能眼镜、智能手表、智能手环等)等。
另外,本文中术语“和/或”,仅仅是一种描述关联对象的关联关系,表示可以存在三种关系,例如,A和/或B,可以表示:单独存在A,同时存在A和B,单独存在B这三种情况。另外,本文中字符“/”,一般表示前后关联对象是一种“或”的关系。
图1为本发明一实施例提供的图像识别方法的流程示意图,如图1所示。
101、获取指定空间的待识别图像。
所谓的指定空间,可以是指储物设备的内部空间,例如,展示柜、开放式货架、冰箱、冷藏柜、等储物设备的内部空间。
102、对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像。
103、对所述至少一个区域图像中每个区域图像进行图像匹配处理,以获得所述每个区域图像所对应的参考图像。
104、根据所述每个区域图像所对应的参考图像的图像信息,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息。
需要说明的是,101~104的执行主体可以为位于本地终端的应用,或者还可以为设置在位于本地终端的应用中的插件或软件开发工具包(Software Development Kit,SDK)等功能单元,或者还可以为位于网络侧服务器中的处理引擎,或者还可以为位于网络侧的分布式系统,本实施例对此不进行特别限定。
可以理解的是,所述应用可以是安装在终端上的本地程序(nativeApp),或者还可以是终端上的浏览器的一个网页程序(webApp),本实施例对此不进行特别限定。
这样,通过获取指定空间的待识别图像,进而对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像,以及对所述至少一个区域图像中每个区域图像进行图像匹配处理,以获得所述每个区域图像所对应的参考图像,使得能够根据所述每个区域图像所对应的参考图像的图像信息,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息,无需人工参与,操作简单,而且正确率高,从而提高了识别的效率和可靠性。
所谓的图像,是指采用一定的图像格式,将图像数据即图像的像素按照一定的方式进行存储,所形成的文件,又可以称为图像文件。
其中,图像的图像格式即图像存储的格式,可以包括但不限于位图(Bitmap,BMP)格式、可移植网络图像格式(Portable Network Graphic Format,PNG)、联合图像专家组(Joint Photographic Experts Group,JPEG)格式、可交换图像文件格式(Exchangeable Image File Format,EXIF),本实施例对此不进行特别限定。
由于指定空间的变化主要是人为操作所导致,例如,操作者打开指
定空间所在储物设备的门体,将某个物品拿走等,因此,可以根据所述指定空间所在储物设备的门体的运动状态,控制摄像装置的开启和关闭,以利用所述摄像装置,拍摄所述指定空间的待识别图像,能够有效降低摄像装置的功率消耗。另外,能够有效降低系统处理所拍摄的待识别图像所需要的处理负担。
可选地,在本实施例的一个可能的实现方式中,在101之前,还可以进一步获取所述指定空间所在储物设备的门体的运动状态,进而,则可以根据所述运动状态,控制摄像装置的开启和关闭,以利用所述摄像装置,拍摄所述指定空间的待识别图像。
具体地,摄像装置可以采用图像传感器,采集指定空间的待识别图像。其中,所述图像传感器可以为电荷耦合元件(Charge Coupled Device,CCD)传感器,或者还可以为金属氧化物半导体元件(Complementary Metal-Oxide Semiconductor,CMOS)传感器,本实施例对此不进行特别限定。此外,根据摄像装置所要拍摄图像内容的位置,可以将摄像装置安装在储物设备的门体上,或者还可以安装在储物设备的体内,此时需要保证储物设备内部的物品不能遮挡摄像装置,本实施例对此不进行特别限定。
在一个具体的实现过程中,具体可以利用设置在所述门体上的传感器装置,获得所述门体的运动参数,例如,加速度、旋转角速度和旋转角度中的至少一项,进而,则可以根据所述运动参数,获得所述门体的运动状态。
例如,可以利用加速度传感器,获得所述门体的加速度,或者还可以利用陀螺仪,获得门体的旋转角速度,或者还可以利用加速度传感器
和陀螺仪,获得门体的旋转角速度和旋转角度,本实施例对此不进行特别限定。
假设门体开启方向为正方向,若旋转角速度为正,且大于或等于预先设置的开启阈值,则可以判断门体为开启;门体开启之后,加速度小于或等于预先设置的静止阈值,则可以判断门体为静止;若旋转角速度为负,且旋转角度小于或等于预先设置的关闭阈值,则可以判断门体为关闭。
那么,当门体静止时,则可以向摄像装置发送开始指令;当门体关闭时,则可以向摄像装置发送停止指令。接收到开始指令之后,摄像装置开始拍摄所述指定空间的待识别图像,直到接收到停止指令为止。
例如,接收到开始指令之后,摄像装置可以以预先设置的拍摄频率,开始对所述指定空间进行连续拍摄处理,以获得所述指定空间在本次拍摄的若干个拍摄图像,将这些拍摄图像进行存储处理。接收到停止指令之后,从所存储的拍摄图像中,选择一个拍摄图像进行策略处理,以确定该拍摄图像是否需要作为所述待识别图像,以供执行101~104。若需要作为所述待识别图像,则可以利用无线通信模块,例如,无线相容性认证(Wireless Fidelity,简称Wi-Fi)模块、全球移动通信系统(Global System for Mobile Communications,GSM)模块、通用分组无线业务(General Packet Radio Service,GPRS)模块等,将所述待识别图像上传到服务器。服务器获取到待识别图像之后,可以对该待识别图像进行图像识别处理,并将处理结果更新到识别平台。
具体来说,所采用的策略处理方法,并不是本申请需要关注的内容,此处不再赘述。进一步地,为了节省存储空间,还可以进一步将没有选
择的其他拍摄图像进行删除处理。
可选地,在本实施例的一个可能的实现方式中,在102中,具体可以利用物品检测器,对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像。
在一个具体的实现过程中,可以预先利用训练样本集,进行训练,以构建所采用的物品检测器。
需要说明的是,训练样本集中所包含的训练样本,可以为经过标注的已知样本,这样,可以直接利用这些已知样本进行训练,以构建物品检测器;或者还可以一部分为经过标注的已知样本,另一部分为没有经过标注的未知样本,那么,则可以先利用已知样本进行训练,以构建初始的物品检测器,然后,再利用初始的物品检测器对未知样本进行预测,以获得检测结果,进而则可以根据未知样本的检测结果,对未知样本进行标注,以形成已知样本,作为新增加的已知样本,利用新增加的已知样本,以及原始的已知样本重新进行训练,以构建新的物品检测器,直到所构建的物品检测器或已知样本满足物品检测器的截止条件为止,如检测准确率大于或等于预先设置的准确率阈值或已知样本的数量大于或等于预先设置的数量阈值等,本实施例对此不进行特别限定。
在该实现方式中,物品检测器可以采用多种方法,获得所述指定空间的至少一个区域图像,本实施例对此不进行特别限定。
一种方法可以是采用基于候选图像集合的检测方法,提取待识别图像的若干个区域的候选图像,对每个候选图像进行检测,从而获得每个候选图像是否为需要检测物品的物品图像。如果候选图像是需要检测物品的物品图像,将该候选图像作为一个所述区域图像。
另一种方法可以是直接检测定位的检测方法,通过直接定位物品所在位置和物品所在区域的方法,检测出待识别图像中物品的位置,从而获得物品所对应的区域图像。
可选地,在本实施例的一个可能的实现方式中,在103中,具体可以利用所述至少一个区域图像中每个区域图像,在预先采集的指定物品的参考图像中进行图像匹配处理,以获得所述每个区域图像所对应的参考图像。
具体地,具体可以采用现有技术中的多种匹配方法,进行图像匹配处理,详细描述可以参见现有技术中的相关内容,此处不再赘述。若匹配到大于或等于预先设置的匹配阈值的参考图像,则可以确定该区域图像中是真实的需要识别的物体。若匹配到小于所述匹配阈值的参考图像,或者未匹配到任何参考图像,则可以确定该区域图像中不是真实的需要识别的物体。
可选地,在本实施例的一个可能的实现方式中,在104中,具体可以根据所述每个区域图像所对应的参考图像的图像信息和该参考图像的识别权重,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息。
在一个具体的实现过程中,具体可以预先对每个参考图像进行标记,以获得每个参考图像的图像信息,例如,物品的名称、产地等基本属性。除了标记参考图像之外,还需要设置每个参考图像的识别权重,以供根据这些参考图像的图像信息进行所匹配的区域图像的识别处理。
假设某个区域图像匹配到三个参考图像,这三个参考图像的识别权重是相等的。一个参考图像的图像信息是北冰洋,另一个参考图像的图
像信息也是北冰洋,还有一个参考图像的图像信息是可口可乐,那么,则可以根据这三个参考图像的图像信息,以及每个参考图像的识别权重,获得该区域图像的物品信息为北冰洋。
本实施例中,通过获取指定空间的待识别图像,进而对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像,以及对所述至少一个区域图像中每个区域图像进行图像匹配处理,以获得所述每个区域图像所对应的参考图像,使得能够根据所述每个区域图像所对应的参考图像的图像信息,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息,无需人工参与,操作简单,而且正确率高,从而提高了识别的效率和可靠性。
另外,采用本发明所提供的技术方案,一旦获取指定空间的待识别图像,即能够自动识别出该待识别图像中的物品信息,能够有效提高识别的效率,而且能够有效提高识别的自动化程度。
另外,采用本发明所提供的技术方案,由于指定空间的变化主要是人为操作所导致,例如,操作者打开指定空间所在储物设备的门体,将某个物品拿走等,因此,可以根据所述指定空间所在储物设备的门体的运动状态,控制摄像装置的开启和关闭,以利用所述摄像装置,拍摄所述指定空间的待识别图像,能够有效降低摄像装置的功率消耗。
需要说明的是,对于前述的各方法实施例,为了简单描述,故将其都表述为一系列的动作组合,但是本领域技术人员应该知悉,本发明并不受所描述的动作顺序的限制,因为依据本发明,某些步骤可以采用其他顺序或者同时进行。其次,本领域技术人员也应该知悉,说明书中所描述的实施例均属于优选实施例,所涉及的动作和模块并不一定是本发
明所必须的。
在上述实施例中,对各个实施例的描述都各有侧重,某个实施例中没有详述的部分,可以参见其他实施例的相关描述。
图2为本发明另一实施例提供的图像识别装置的结构示意图,如图2所示。本实施例的图像识别装置可以包括获取单元21、分割单元22、匹配单元23和识别单元24。其中,获取单元21,用于获取指定空间的待识别图像;分割单元22,用于对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像;匹配单元23,用于对所述至少一个区域图像中每个区域图像进行图像匹配处理,以获得所述每个区域图像所对应的参考图像;识别单元24,用于根据所述每个区域图像所对应的参考图像的图像信息,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息。
所谓的指定空间,可以是指储物设备的内部空间,例如,展示柜、开放式货架、冰箱、冷藏柜、等储物设备的内部空间。
需要说明的是,本实施例所提供的图像识别装置可以为位于本地终端的应用,或者还可以为设置在位于本地终端的应用中的插件或软件开发工具包(Software Development Kit,SDK)等功能单元,或者还可以为位于网络侧服务器中的处理引擎,或者还可以为位于网络侧的分布式系统,本实施例对此不进行特别限定。
可以理解的是,所述应用可以是安装在终端上的本地程序(nativeApp),或者还可以是终端上的浏览器的一个网页程序(webApp),本实施例对此不进行特别限定。
可选地,在本实施例的一个可能的实现方式中,所述分割单元22,
具体可以用于利用物品检测器,对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像。
可选地,在本实施例的一个可能的实现方式中,所述匹配单元23,具体可以用于利用所述至少一个区域图像中每个区域图像,在预先采集的指定物品的参考图像中进行图像匹配处理,以获得所述每个区域图像所对应的参考图像。
可选地,在本实施例的一个可能的实现方式中,所述识别单元24,具体可以用于根据所述每个区域图像所对应的参考图像的图像信息和该参考图像的识别权重,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息。
可选地,在本实施例的一个可能的实现方式中,如图3所示,本实施例所提供的图像识别装置还可以进一步包括控制单元31,可以用于获取所述指定空间所在储物设备的门体的运动状态;以及根据所述运动状态,控制摄像装置的开启和关闭,以利用所述摄像装置,拍摄所述指定空间的待识别图像。
具体地,所述控制单元31,具体可以用于利用传感器装置,获得所述门体的运动参数;以及根据所述运动参数,获得所述门体的运动状态。
需要说明的是,图1对应的实施例中方法,可以由本实施例提供的图像识别装置实现。详细描述可以参见图1对应的实施例中的相关内容,此处不再赘述。
本实施例中,通过获取单元获取指定空间的待识别图像,进而由分割单元对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像,以及由匹配单元对所述至少一个区域图像中每个区域
图像进行图像匹配处理,以获得所述每个区域图像所对应的参考图像,使得识别单元能够根据所述每个区域图像所对应的参考图像的图像信息,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息,无需人工参与,操作简单,而且正确率高,从而提高了识别的效率和可靠性。
另外,采用本发明所提供的技术方案,一旦获取指定空间的待识别图像,即能够自动识别出该待识别图像中的物品信息,能够有效提高识别的效率,而且能够有效提高识别的自动化程度。
另外,采用本发明所提供的技术方案,由于指定空间的变化主要是人为操作所导致,例如,操作者打开指定空间所在储物设备的门体,将某个物品拿走等,因此,可以根据所述指定空间所在储物设备的门体的运动状态,控制摄像装置的开启和关闭,以利用所述摄像装置,拍摄所述指定空间的待识别图像,能够有效降低摄像装置的功率消耗。
所属领域的技术人员可以清楚地了解到,为描述的方便和简洁,上述描述的系统,装置和单元的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
在本发明所提供的几个实施例中,应该理解到,所揭露的系统,装置和方法,可以通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,装置或单元的间接耦合或通信连接,可以是电性,机械或其它的形
式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本发明各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现,也可以采用硬件加软件功能单元的形式实现。
上述以软件功能单元的形式实现的集成的单元,可以存储在一个计算机可读取存储介质中。上述软件功能单元存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)或处理器(processor)执行本发明各个实施例所述方法的部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
最后应说明的是:以上实施例仅用以说明本发明的技术方案,而非对其限制;尽管参照前述实施例对本发明进行了详细的说明,本领域的普通技术人员应当理解:其依然可以对前述各实施例所记载的技术方案进行修改,或者对其中部分技术特征进行等同替换;而这些修改或者替换,并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围。
Claims (16)
- 一种图像识别方法,其特征在于,包括:获取指定空间的待识别图像;对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像;对所述至少一个区域图像中每个区域图像进行图像匹配处理,以获得所述每个区域图像所对应的参考图像;根据所述每个区域图像所对应的参考图像的图像信息,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息。
- 根据权利要求1所述的方法,其特征在于,所述指定空间包括储物设备的内部空间。
- 根据权利要求1或2所述的方法,其特征在于,所述对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像,包括:利用物品检测器,对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像。
- 根据权利要求1~3任一权利要求所述的方法,其特征在于,所述对所述至少一个区域图像中每个区域图像进行图像匹配处理,以获得所述每个区域图像所对应的参考图像,包括:利用所述至少一个区域图像中每个区域图像,在预先采集的指定物品的参考图像中进行图像匹配处理,以获得所述每个区域图像所对应的参考图像。
- 根据权利要求1~4任一权利要求所述的方法,其特征在于,所 述根据所述每个区域图像所对应的参考图像的图像信息,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息,包括:根据所述每个区域图像所对应的参考图像的图像信息和该参考图像的识别权重,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息。
- 根据权利要求1~5任一权利要求所述的方法,其特征在于,所述获取指定空间的待识别图像之前,还包括:获取所述指定空间所在储物设备的门体的运动状态;以及根据所述运动状态,控制摄像装置的开启和关闭,以利用所述摄像装置,拍摄所述指定空间的待识别图像。
- 根据权利要求6所述的方法,其特征在于,所述获取所述指定空间所在储物设备的门体的运动状态,包括:利用传感器装置,获得所述门体的运动参数;以及根据所述运动参数,获得所述门体的运动状态。
- 一种图像识别装置,其特征在于,包括:获取单元,用于获取指定空间的待识别图像;分割单元,用于对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像;匹配单元,用于对所述至少一个区域图像中每个区域图像进行图像匹配处理,以获得所述每个区域图像所对应的参考图像;识别单元,用于根据所述每个区域图像所对应的参考图像的图像信息,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息。
- 根据权利要求8所述的装置,其特征在于,所述指定空间包括储物设备的内部空间。
- 根据权利要求8或9所述的装置,其特征在于,所述分割单元,具体用于利用物品检测器,对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像。
- 根据权利要求8~10任一权利要求所述的装置,其特征在于,所述匹配单元,具体用于利用所述至少一个区域图像中每个区域图像,在预先采集的指定物品的参考图像中进行图像匹配处理,以获得所述每个区域图像所对应的参考图像。
- 根据权利要求8~11任一权利要求所述的装置,其特征在于,所述识别单元,具体用于根据所述每个区域图像所对应的参考图像的图像信息和该参考图像的识别权重,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息。
- 根据权利要求8~12任一权利要求所述的装置,其特征在于,所述装置还包括控制单元,用于获取所述指定空间所在储物设备的门体的运动状态;以及根据所述运动状态,控制摄像装置的开启和关闭,以利用所述摄像装置,拍摄所述指定空间的待识别图像。
- 根据权利要求13所述的装置,其特征在于,所述控制单元,具体用于利用传感器装置,获得所述门体的运动参数;以及根据所述运动参数,获得所述门体的运动状态。
- 一种设备,包括:一个或者多个处理器;存储器;一个或者多个程序,所述一个或者多个程序存储在所述存储器中,当被所述一个或者多个处理器执行时:获取指定空间的待识别图像;对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像;对所述至少一个区域图像中每个区域图像进行图像匹配处理,以获得所述每个区域图像所对应的参考图像;根据所述每个区域图像所对应的参考图像的图像信息,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息。
- 一种非易失性计算机存储介质,所述非易失性计算机存储介质存储有一个或者多个程序,当所述一个或者多个程序被一个设备执行时,使得所述设备:获取指定空间的待识别图像;对所述待识别图像进行图像分割处理,以获得所述指定空间的至少一个区域图像;对所述至少一个区域图像中每个区域图像进行图像匹配处理,以获得所述每个区域图像所对应的参考图像;根据所述每个区域图像所对应的参考图像的图像信息,对所述每个区域图像进行识别处理,以获得所述每个区域图像的物品信息。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US16/081,329 US10706314B2 (en) | 2016-03-23 | 2016-05-23 | Image recognition method, apparatus and device, and non-volatile computer storage medium |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610170233.5 | 2016-03-23 | ||
CN201610170233.5A CN105868772A (zh) | 2016-03-23 | 2016-03-23 | 图像识别方法及装置 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2017161665A1 true WO2017161665A1 (zh) | 2017-09-28 |
Family
ID=56625147
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2016/082970 WO2017161665A1 (zh) | 2016-03-23 | 2016-05-23 | 图像识别方法、装置、设备及非易失性计算机存储介质 |
Country Status (3)
Country | Link |
---|---|
US (1) | US10706314B2 (zh) |
CN (1) | CN105868772A (zh) |
WO (1) | WO2017161665A1 (zh) |
Families Citing this family (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018105171A1 (ja) * | 2016-12-06 | 2018-06-14 | コニカミノルタ株式会社 | 画像認識システム及び画像認識方法 |
CN108038411A (zh) * | 2017-10-31 | 2018-05-15 | 珠海格力电器股份有限公司 | 账单生成方法、装置及冰箱 |
CN108549870B (zh) * | 2018-04-16 | 2022-06-10 | 图麟信息科技(深圳)有限公司 | 一种对物品陈列进行鉴别的方法及装置 |
CN110765825A (zh) | 2018-07-27 | 2020-02-07 | 北京京东尚科信息技术有限公司 | 物品摆放状态获取方法和系统 |
CN109190676B (zh) * | 2018-08-06 | 2022-11-08 | 百度在线网络技术(北京)有限公司 | 一种用于图像识别的模型训练方法、装置、设备及存储介质 |
CN109448047A (zh) * | 2018-09-18 | 2019-03-08 | 北京无线体育俱乐部有限公司 | 货架分布图生成方法、信息获取方法、装置及系统 |
CN111102966A (zh) * | 2018-10-29 | 2020-05-05 | 所罗门股份有限公司 | 自动采集设备状态的方法 |
CN109668852B (zh) * | 2018-12-25 | 2021-11-02 | Oppo广东移动通信有限公司 | 电子设备、信息推送方法及相关产品 |
CN111666959A (zh) * | 2019-03-06 | 2020-09-15 | 西安邮电大学 | 矢量图像匹配方法及装置 |
CN110705511A (zh) * | 2019-10-16 | 2020-01-17 | 北京字节跳动网络技术有限公司 | 模糊图像的识别方法、装置、设备及存储介质 |
CN110765525B (zh) * | 2019-10-18 | 2023-11-10 | Oppo广东移动通信有限公司 | 生成场景图片的方法、装置、电子设备及介质 |
WO2021092883A1 (zh) * | 2019-11-15 | 2021-05-20 | 深圳市微蓝智能科技有限公司 | 物品管理方法、终端装置和物品管理设备及存储介质 |
CN113129303B (zh) * | 2021-05-18 | 2022-01-18 | 广州市吉华勘测股份有限公司 | 一种巡检图片自动标注方法、装置、存储介质及电子设备 |
CN114862867A (zh) * | 2021-12-27 | 2022-08-05 | 青岛海尔电冰箱有限公司 | 冰箱内物品信息识别方法、冰箱和计算机存储介质 |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050196046A1 (en) * | 2004-01-30 | 2005-09-08 | Evolution Robotics, Inc. | Vision-enabled household appliances |
CN102982332A (zh) * | 2012-09-29 | 2013-03-20 | 顾坚敏 | 基于云处理方式的零售终端货架影像智能分析系统 |
CN103793192A (zh) * | 2014-02-25 | 2014-05-14 | 四川长虹电器股份有限公司 | 一种信息输出方法、冰箱及服务器 |
CN104061748A (zh) * | 2013-03-18 | 2014-09-24 | 海尔集团公司 | 一种冰箱内食品存储位置记录方法、装置、终端及冰箱 |
CN104990325A (zh) * | 2015-06-17 | 2015-10-21 | 张梅云 | 一种智能储物装置 |
CN105180590A (zh) * | 2015-10-28 | 2015-12-23 | 合肥美的电冰箱有限公司 | 一种基于指纹识别的冰箱管理方法、管理系统和冰箱 |
Family Cites Families (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5546475A (en) * | 1994-04-29 | 1996-08-13 | International Business Machines Corporation | Produce recognition system |
US5638465A (en) * | 1994-06-14 | 1997-06-10 | Nippon Telegraph And Telephone Corporation | Image inspection/recognition method, method of generating reference data for use therein, and apparatuses therefor |
AUPQ212499A0 (en) * | 1999-08-10 | 1999-09-02 | Ajax Cooke Pty Ltd | Item recognition method and apparatus |
US7693757B2 (en) * | 2006-09-21 | 2010-04-06 | International Business Machines Corporation | System and method for performing inventory using a mobile inventory robot |
CN101542531B (zh) * | 2007-06-14 | 2012-12-19 | 松下电器产业株式会社 | 图像识别装置及图像识别方法 |
US8170280B2 (en) * | 2007-12-03 | 2012-05-01 | Digital Smiths, Inc. | Integrated systems and methods for video-based object modeling, recognition, and tracking |
CN101853387A (zh) * | 2010-04-02 | 2010-10-06 | 北京物资学院 | 立体仓库货物盘点方法及系统 |
CN102063616A (zh) * | 2010-12-30 | 2011-05-18 | 上海电机学院 | 一种基于图像特征匹配的商品自动识别系统及方法 |
US9367761B2 (en) * | 2013-02-15 | 2016-06-14 | Samsung Electronics Co., Ltd. | Electronic device and object recognition method in electronic device |
CN103150631B (zh) * | 2013-03-11 | 2016-11-09 | 苏州斯普锐智能系统有限公司 | 基于图像匹配的自动物品管理方法及其系统 |
US9349113B2 (en) * | 2013-03-26 | 2016-05-24 | 3 Strike, Llc | Storage container with inventory control |
US9098825B2 (en) * | 2013-03-26 | 2015-08-04 | Leonard Bashkin | Storage container with inventory control |
CN103617420A (zh) * | 2013-11-27 | 2014-03-05 | 上海电机学院 | 基于图像特征匹配的商品快速识别方法及系统 |
US9720934B1 (en) * | 2014-03-13 | 2017-08-01 | A9.Com, Inc. | Object recognition of feature-sparse or texture-limited subject matter |
US9240077B1 (en) * | 2014-03-19 | 2016-01-19 | A9.Com, Inc. | Real-time visual effects for a live camera view |
JP6471934B2 (ja) * | 2014-06-12 | 2019-02-20 | パナソニックIpマネジメント株式会社 | 画像認識方法、カメラシステム |
CN204002281U (zh) * | 2014-08-28 | 2014-12-10 | 东莞市邦泽电子有限公司 | 一种具有远程管理功能的保险柜 |
CN204795393U (zh) * | 2015-06-29 | 2015-11-18 | 江苏龙芯梦兰信息安全技术有限公司 | 保险柜新型智能监控终端 |
CN105118123B (zh) * | 2015-08-26 | 2017-06-23 | 王伟 | 一种智能门装置和智能小区系统 |
-
2016
- 2016-03-23 CN CN201610170233.5A patent/CN105868772A/zh active Pending
- 2016-05-23 US US16/081,329 patent/US10706314B2/en active Active
- 2016-05-23 WO PCT/CN2016/082970 patent/WO2017161665A1/zh active Application Filing
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050196046A1 (en) * | 2004-01-30 | 2005-09-08 | Evolution Robotics, Inc. | Vision-enabled household appliances |
CN102982332A (zh) * | 2012-09-29 | 2013-03-20 | 顾坚敏 | 基于云处理方式的零售终端货架影像智能分析系统 |
CN104061748A (zh) * | 2013-03-18 | 2014-09-24 | 海尔集团公司 | 一种冰箱内食品存储位置记录方法、装置、终端及冰箱 |
CN103793192A (zh) * | 2014-02-25 | 2014-05-14 | 四川长虹电器股份有限公司 | 一种信息输出方法、冰箱及服务器 |
CN104990325A (zh) * | 2015-06-17 | 2015-10-21 | 张梅云 | 一种智能储物装置 |
CN105180590A (zh) * | 2015-10-28 | 2015-12-23 | 合肥美的电冰箱有限公司 | 一种基于指纹识别的冰箱管理方法、管理系统和冰箱 |
Also Published As
Publication number | Publication date |
---|---|
US10706314B2 (en) | 2020-07-07 |
US20190019053A1 (en) | 2019-01-17 |
CN105868772A (zh) | 2016-08-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2017161665A1 (zh) | 图像识别方法、装置、设备及非易失性计算机存储介质 | |
US20180054564A1 (en) | Apparatus and method for providing user's emotional information in electronic device | |
CN106296617B (zh) | 人脸图像的处理方法及装置 | |
CN106031154B (zh) | 处理图像的方法和用于其的电子装置 | |
CN105160327A (zh) | 建筑物识别方法和装置 | |
CN106254807B (zh) | 提取静止图像的电子设备和方法 | |
CN101609660A (zh) | 数字相框、信息处理系统和控制方法 | |
US9691180B2 (en) | Determination of augmented reality information | |
WO2018171047A1 (zh) | 一种拍摄引导方法、设备及系统 | |
KR102223308B1 (ko) | 영상처리 방법 및 이를 구현한 전자 장치 | |
CN104731880A (zh) | 图片排序方法和装置 | |
CN112115894A (zh) | 手部关键点检测模型的训练方法、装置及电子设备 | |
EP2960816A1 (en) | Method and apparatus for managing data | |
KR20150141426A (ko) | 전자 장치 및 전자 장치에서의 이미지 처리 방법 | |
US9547669B2 (en) | Performing a visual search using a rectified image | |
EP2659429A1 (en) | Methods, apparatuses and computer program products for efficiently recognizing faces of images associated with various illumination conditions | |
JP2015090593A (ja) | 情報処理装置、情報処理方法および情報処理システム | |
CN105472337B (zh) | 图像监控方法及装置 | |
CN113269730A (zh) | 图像处理方法、装置、计算机设备及存储介质 | |
US10108861B2 (en) | Systems and methods of providing content differentiation between thumbnails | |
CN114341946A (zh) | 识别方法、装置、电子设备及存储介质 | |
JP6112346B2 (ja) | 情報収集システム、プログラムおよび情報収集方法 | |
CN104881149A (zh) | 基于视频流的输入方法及装置 | |
CN110781796B (zh) | 一种标注方法、装置及电子设备 | |
WO2017099935A1 (en) | Motion detection of object |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16895017 Country of ref document: EP Kind code of ref document: A1 |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 16895017 Country of ref document: EP Kind code of ref document: A1 |