WO2019000313A1 - 一种检测方法、检测设备以及飞行器 - Google Patents

一种检测方法、检测设备以及飞行器 Download PDF

Info

Publication number
WO2019000313A1
WO2019000313A1 PCT/CN2017/090778 CN2017090778W WO2019000313A1 WO 2019000313 A1 WO2019000313 A1 WO 2019000313A1 CN 2017090778 W CN2017090778 W CN 2017090778W WO 2019000313 A1 WO2019000313 A1 WO 2019000313A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
acquisition component
polarization direction
view
field
Prior art date
Application number
PCT/CN2017/090778
Other languages
English (en)
French (fr)
Inventor
崔健
封旭阳
赵丛
Original Assignee
深圳市大疆创新科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳市大疆创新科技有限公司 filed Critical 深圳市大疆创新科技有限公司
Priority to CN201780005493.XA priority Critical patent/CN108701356A/zh
Priority to PCT/CN2017/090778 priority patent/WO2019000313A1/zh
Publication of WO2019000313A1 publication Critical patent/WO2019000313A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G02—OPTICS
    • G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
    • G02B27/00—Optical systems or apparatus not provided for by any of the groups G02B1/00 - G02B26/00, G02B30/00
    • G02B27/28—Optical systems or apparatus not provided for by any of the groups G02B1/00 - G02B26/00, G02B30/00 for polarising
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10032—Satellite or aerial image; Remote sensing

Definitions

  • the embodiments of the present application relate to the field of the technology of the drone, and in particular, to a detecting method, a detecting device, and an aircraft.
  • UAVs In the flight environment of the drone, there are often specular objects, and the mirror objects pose a great safety hazard to the flight safety of the drone.
  • UAVs currently mainly use vision sensors to collect images of obstacles to achieve obstacle avoidance.
  • vision sensors does not capture images of specular objects well, because the images acquired by visual sensors are often reflected by specular objects.
  • specular objects The image formed by light, therefore, can not accurately analyze the mirror object, and it is easy to produce false positives.
  • specular objects There is currently a lack of detection of specular objects to avoid safe flight accidents of drones that may be caused by specular objects.
  • Embodiments of the present invention provide a detection method, a detection device, and an aircraft for detecting a mirror object easily, quickly, and accurately.
  • the first aspect of the present invention provides a detection method, which may include:
  • the polarization component is disposed on the image acquisition component, and the polarization direction of the polarization component when acquiring the first image is different from the polarization direction of the polarization component when acquiring the second image;
  • a second aspect of the present invention provides a detecting apparatus, which may include:
  • the image acquisition component is used to:
  • a third aspect of the present invention provides an aircraft, which may include:
  • a power system disposed on the fuselage for providing flight power to the aircraft
  • a detection device as in the second aspect is a detection device as in the second aspect.
  • the first image and the second image can be acquired in different polarization directions, and then can be determined according to the first image and the second image. Whether there is a specular object within the field of view of the image acquisition component.
  • the polarization direction of the polarization element is different from the polarization direction of the polarization element when the first image is acquired.
  • the incoming light entering the image acquisition component when acquiring the first image Different from the incoming light entering the image capturing component when acquiring the second image, there is a significant difference between the first image and the second image, so that the first image and the second image can be detected easily, quickly, and accurately. Whether there is a specular object within the field of view of the image acquisition component.
  • FIG. 1 is a schematic diagram of an embodiment of a detection method according to an embodiment of the present invention.
  • FIG. 2 is a schematic diagram of light intensity distribution of a first image according to an embodiment of the present invention.
  • FIG. 3 is a schematic diagram of light intensity distribution of a second image according to an embodiment of the present invention.
  • FIG. 4 is a schematic diagram of another embodiment of a detection method according to an embodiment of the present invention.
  • FIG. 5 is a schematic diagram of a first preset process of a first image and a second image according to an embodiment of the present invention
  • FIG. 6 is a schematic diagram of a second preset process of a first image and a second image according to an embodiment of the present invention
  • FIG. 7 is a schematic diagram of another embodiment of a detection method according to an embodiment of the present invention.
  • FIG. 8 is a schematic diagram of an embodiment of a detecting device according to an embodiment of the present invention.
  • Embodiments of the present invention provide a detection method, a detection device, and an aircraft for detecting a mirror object easily, quickly, and accurately.
  • an embodiment of the detection method in the embodiment of the present invention includes:
  • the method may be performed by a detecting device, wherein the detecting device includes an image capturing component, and the image capturing component is provided with a polarizing component.
  • the image capturing component performs image capturing, different polarization directions of the polarizing component may be respectively used.
  • the acquisition of the first image and the second image is performed, that is, the polarization direction of the polarization element when acquiring the first image is different from the polarization direction of the polarization element when the second image is acquired.
  • the reflected light will have a polarization effect, that is, generally partially polarized light (in particular, linearly polarized light).
  • the intensity of the partially polarized light is asymmetric with respect to the direction of propagation relative to the intensity of the axis of the natural light, and the distribution of the intensity of the partially polarized light is non-axisymmetric.
  • the polarizing element is provided with a special direction, that is, a polarization direction, and only light having a vibration direction parallel to the polarization direction can pass through the polarizing element, and therefore, since the polarization directions of the polarizing elements are different when the first image and the second image are acquired, the first The image and the second image will have different characteristics, such as different light intensity distributions.
  • the source of the light collected by the image capturing component of the first image and the second image should be consistent, and at the same time, at least the intersection of the image content of the first image and the second image can be combined to combine the first image and the second image pair.
  • Objects within the field of view of the image acquisition component are correctly and efficiently detected.
  • the image content of the first image may completely coincide with the image content of the second image, that is, the first image and the second image are identical except for being acquired by different polarization directions.
  • the specific manner of acquiring the first image and the second image collected by the image capturing component may be:
  • a second image is acquired by a second one of the two image sensing devices.
  • the image acquisition component can include two image sensing devices, wherein each of the image sensing devices is provided with a polarizing element.
  • the image sensing device may include an optical lens, that is, the image capturing component may include two optical lenses, such as a binocular camera, and the surfaces of the two optical lenses may be respectively covered with a layer such as a polarizing plate (polarizing element) and two polarizing plates.
  • the polarizing angles are different, so that the polarization directions of the two polarizing plates are inconsistent, so that when two optical lenses are taken, the polarization directions of the light entering the two optical lenses are different, thereby achieving different polarizations in the polarizing plates.
  • the direction acquires the first image and the second image, respectively.
  • the image capturing component in this embodiment may further include two or more image sensing devices, wherein each of the at least two image sensing devices may be configured on the image sensing device.
  • the image sensing device using the polarizing element is used The more the area of the object in the field of view of the image acquisition component is, the higher the accuracy is, which is not limited herein.
  • the image acquisition component may be, for example, a monocular camera, and the monocular camera may be configured with a polarization element, and the polarization angle of the polarization component may be adjusted to adjust to different polarizations.
  • Direction of vibration In practical applications.
  • the polarization direction of the polarizing element may be first set to a first polarization direction, and the first image may be acquired by the image acquisition component in the first polarization direction, and then the polarization direction of the polarization element may be adjusted from the first polarization direction to the second direction.
  • a polarization direction, and a second image is acquired by the image acquisition component in the second polarization direction.
  • the polarizing element may be a liquid crystal device, and the liquid crystal device is a device capable of controlling the polarization direction by an electrical signal. Therefore, in this embodiment, the polarization direction can be switched by using the energization condition of the liquid crystal device. For example, after the first image is acquired by the polarization lateral direction, the polarization longitudinal direction can be switched and the second image is acquired, so that two liquid crystal devices can be obtained. Images acquired in different polarization directions.
  • the polarizing element in this embodiment may be other types of polarizing elements in addition to the liquid crystal device described above, as long as the polarization direction can be switched. This is not limited here.
  • the image capturing component may also include more than one image sensing device, wherein at least one of the image sensing devices is provided with a polarizing component capable of switching a polarization direction, such as a liquid crystal device, and the configuration used is The more image sensing devices of the polarizing element, the larger the area of the object within the field of view of the image capturing component, the higher the accuracy.
  • the polarization direction of the polarizing element is different from the polarization direction of the polarizing element when the first image is obtained, which is not limited herein.
  • the detecting device where the image capturing component is located may be disposed on a moving object, such as a moving object such as an aircraft or a remote control car, so as to better assist the aircraft, the remote control car, and the like to avoid the mirror object, and complete the obstacle avoidance operation. Prevent the occurrence of security incidents.
  • whether the mirror object exists in the field of view of the image capturing component may be determined according to the first image and the second image.
  • the reflected light and the refracted light are no longer natural light, but partially polarized light, and in the reflected light, the light perpendicular to the incident surface vibrates more than the parallel vibration, and the refracted light is the opposite.
  • the degree of polarization is related to the angle of incidence. When the angle of incidence is equal to the Brewster angle, the reflected light becomes linearly polarized light that is perpendicular to the plane of incidence.
  • the first image and the second image may have a significant difference.
  • a mirror object in the field of view of the image acquisition component, and two image sensing devices are arranged on the image acquisition component, that is, a binocular camera.
  • each optical camera of the binocular camera Polarizers with eccentric angles perpendicular to each other can be respectively arranged.
  • the binocular camera can obtain the first image and the second image by using two polarization directions of the polarizing plate. image. As shown in FIG. 2 and FIG.
  • the first image acquired in the first polarization direction of one polarizer can be obtained (eg, 2), and a second image acquired in the second polarization direction of the other polarizer (FIG. 3), the light intensity distribution characteristics of the first image and the second image are inconsistent.
  • the intensity distribution characteristics between the first image and the second image will be the same. It can be seen that, according to the acquired first image and second image, it can be determined whether a specular object exists in the field of view of the image capturing component.
  • the mirror object may include at least one of a water surface, a glass, a mirror, and a ceramic tile.
  • the first image and the second image may be acquired in different polarization directions by using the polarization component on the image acquisition component, and then the field of view of the image acquisition component may be determined according to the first image and the second image. Whether there is a specular object.
  • the polarization direction of the polarization element is different from the polarization direction of the polarization element when the first image is acquired.
  • the incoming light entering the image acquisition component when acquiring the first image Different from the incoming light entering the image capturing component when acquiring the second image, there is a significant difference between the first image and the second image, so that the first image and the second image can be detected easily, quickly, and accurately. Whether there is a specular object within the field of view of the image acquisition component.
  • the manner of determining whether there is a specular object in the field of view of the image capturing component may be different, and the following is specifically described:
  • another embodiment of the detecting method in the implementation of the present invention includes:
  • Step 401 in this embodiment is the same as step 101 in the embodiment shown in FIG. 1, and details are not described herein again.
  • the first image and the second image may be subjected to preset processing to obtain the third image.
  • the specific manner of performing preset processing on the first image and the second image to obtain the third image may be:
  • the pixel value of each pixel in the first image is divided into the pixel value of the corresponding pixel in the second image, and the image obtained after the division processing is determined as the third image.
  • an image may include a plurality of pixels, and the value of each pixel may reflect the brightness of the image of the point. Then, by determining the pixel values of each pixel in the first image and the second image, respectively, the brightness of the corresponding position in the first image and the second image can be determined.
  • the pixel values of each pixel point in the first image and the pixel values of the corresponding pixel points in the second image may be subtracted to The image obtained after the subtraction process is a third image, or the pixel value of each pixel in the first image may be divided into the pixel value of the corresponding pixel in the second image, and the image obtained after the division process is obtained. For the third image. By acquiring the third image, it is possible to accurately display the difference between the pixel value of the pixel point in the first image and the pixel value of the corresponding pixel point in the second image.
  • the specific manner of performing the preset processing on the first image and the second image to obtain the third image in addition to the several examples described above, in the practical application, other algorithms may also be adopted, as long as The acquired third image can be used to determine whether a specular object exists in the field of view of the image capturing component, which is not limited herein.
  • whether the mirror object exists in the field of view of the image capturing component may be determined according to the third image.
  • the first image and the second image are subjected to a preset process to obtain a pattern distribution caused by the light intensity distribution feature after the third image is acquired Will change overall, such as the third image relative to the first image or the second image, the third image Like darkening the whole.
  • the first image and the second image are subjected to preset processing to obtain the third image, and the pattern distribution of the third image due to the light intensity distribution feature is not only
  • the brightness will change and the shape will change, such as the first image (Fig. 2) and the second image (Fig. 3) described in step 102 of the embodiment shown in Fig.
  • the specific manner of determining whether there is a specular object in the field of view of the image acquisition component according to the third image may be:
  • the water surface is used as a mirror object to be schematically illustrated.
  • the preset image needs to be determined, wherein the specific process may be: using The image acquisition component described above captures the water surface, and acquires a plurality of first standard images A1, A2, A3, A4, and A5 including a water surface, wherein a polarization direction of the polarization element corresponding to the first standard image is a first polarization direction. And acquiring a plurality of second standard images B1, B2, B3, B4, and B5 including a water surface, wherein a polarization direction of the polarizing element corresponding to the second standard image is a second polarization direction.
  • A1 and B1, A2 and B2, A3 and B3, A4 and B4, A5 and B5 may be pre-processed, such as subtracting or dividing the pixel values of the corresponding pixel points to obtain 5 presets.
  • the five preset processed images are averaged to obtain a preset image for detecting whether there is a water surface in the field of view of the image capturing component. Therefore, in this embodiment, after the third image is acquired, the correlation between the third image and the preset image of the water surface may be determined, and when the correlation between the third image and the preset image of the water surface is greater than a preset threshold, the determination may be determined. There is a water surface in the field of view of the image acquisition component. Conversely, it can be determined that there is no water surface within the field of view of the image acquisition component.
  • the number of the first standard image and the second standard image of the water surface in the embodiment is also Others can be other values.
  • the above is only an example, and the preset image is acquired in addition to the above description. In practical applications, other methods may be used as long as the correlation for detecting the correlation of the third image can be obtained. Set the image, which is not limited here.
  • preset images of different mirror objects can be classified and stored, and at the same time, when detecting the correlation of the third image preset images, the third image and corresponding can be respectively determined for different mirror objects.
  • the correlation of the preset image if there is a preset image of water surface, glass, and mirror surface, three correlations can be obtained, and whether the mirror object is present in the field of view of the image acquisition component can be determined according to the three correlations.
  • the first operation model needs to be trained first, wherein the first operation model may be a neural network, a classifier, or the like.
  • the first operation model may be a neural network, a classifier, or the like.
  • A1 and B1, A2 and B2, A3 and B3, A4 and B4, A5 and B5 are pre-processed, such as subtracting or dividing the pixel values of corresponding pixels to obtain 5 sheets.
  • the five preset processed images can be used as training samples to train the first computing model.
  • the first operational model after training can be used.
  • the third image is detected, and the trained first computing model can output the recognition result.
  • the recognition result can be a confidence level. When the confidence level is higher than a preset threshold, it can be determined that the field of view of the image acquisition component is in the surface. .
  • the training sample includes a standard image of the water surface after the preset processing, and may include a standard image of the other mirror object after the preset processing, and may further include a non-mirror object after the preset processing.
  • Standard image to facilitate distinguishing between different specular objects and feature information of specular objects and non-specular objects, such as regular stripe information of tiles, so that the first operational model after training can accurately mirror objects according to the input third image. Judging and further identifying what the specular object is.
  • the specific manner of determining whether there is a specular object in the field of view of the image capturing component according to the third image may be combined in other ways in practical applications. It can be used alone, as long as it can determine whether there is a specular object in the field of view of the image capturing component according to the third image, which is not limited herein.
  • another embodiment of the detecting method in the implementation of the present invention includes:
  • Step 701 in this embodiment is the same as step 101 in the embodiment shown in FIG. Said.
  • the first image and the second image may be input into a preset second computing model, and the image capturing component is determined according to the recognition result output by the second computing model. Whether there is a specular object within the field of view.
  • the second operation model needs to be trained first, and the second operation model may be a neural network, a classifier, or the like.
  • the water surface is schematically illustrated as a mirror object.
  • the water surface is photographed by using the image acquisition component as described above, and a plurality of first standard images including the water surface are obtained, wherein the first standard image corresponds to the polarization.
  • the polarization direction of the component is a first polarization direction
  • a plurality of second standard images including a water surface are obtained, wherein a polarization direction of the polarization element corresponding to the first standard image is a second polarization direction, and the first standard image and the second standard image are used.
  • the second operational model is trained. After the training of the second operation model is completed, the first image and the second image may be detected by using the second operation model after training, and the second operation model may output the recognition result, for example, the recognition result may be a confidence degree, when the confidence is satisfied When the degree is higher than the preset threshold, it can be determined that the field of view of the image acquisition component is in the water surface.
  • the training sample may include standard images of other mirror objects, and may include standard images of non-mirror objects, in order to distinguish different mirror objects and mirror objects.
  • the feature information of the mirror object such as the regular stripe information of the tile, enables the second computing model after the training to accurately judge the specular object according to the input first image and the second image, and can further identify the object of the mirror object.
  • the present invention only exemplifies the method for detecting a mirror object by using the embodiment shown in FIG. 4 and the embodiment shown in FIG. 7. In practical applications, other methods may be adopted as long as the first image can be used. And the second image determines whether a mirror object exists in the field of view of the image capturing component, which is not limited herein.
  • an embodiment of the detection device in the embodiment of the present invention includes:
  • the image acquisition component 801 is configured to:
  • the processor 802 is configured to:
  • the image acquisition component includes two image sensing devices, wherein each of the image sensing devices is configured with a polarizing element;
  • the image acquisition component 801 can be further specifically used to:
  • a second image is acquired by a second one of the two image sensing devices.
  • the processor 802 may be further specifically used to:
  • the image capturing component 801 can be further specifically configured to: collect the first image
  • the processor 802 can be further specifically used for:
  • the image capturing component 801 can be further specifically configured to: collect the second image.
  • the processor 802 may be further specifically used to:
  • a specular object is determined within the field of view of the image acquisition component based on the third image.
  • the processor 802 may be further specifically used to:
  • the pixel value of each pixel in the first image and the pixel value of the corresponding pixel in the second image are subtracted, and the image obtained after the subtraction process is determined as the third image.
  • the processor 802 may be further specifically used to:
  • the pixel value of each pixel in the first image is divided into the pixel value of the corresponding pixel in the second image, and the image obtained after the division processing is determined as the third image.
  • the processor 802 may be further specifically used to:
  • a correlation between the third image and the preset image is detected, and whether a specular object exists within the field of view of the image capturing component is determined according to the correlation.
  • the processor 802 may be further specifically used to:
  • the processor 802 may be further specifically used to:
  • the embodiment may further include a memory 803 on which a computer program may be stored, so that when the processor 802 is configured to execute a computer program stored in the memory 803, the steps described above may be implemented.
  • the detecting device may include, but is not limited to, an image capturing component, a processor, and a memory. It will be understood by those skilled in the art that the schematic diagram of FIG. 8 is merely an example of a detecting device, does not constitute a limitation of the detecting device, may include more or less components than those illustrated, or combine some components, or different components, For example, the detecting device may further include an input/output device, a network access device, and the like.
  • the processor may be a central processing unit (CPU), or may be other general-purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), and an off-the-shelf device.
  • DSP digital signal processor
  • ASIC application specific integrated circuit
  • FPGA Field-Programmable Gate Array
  • the general purpose processor may be a microprocessor or the processor or any conventional processor or the like, which is the control center of the detecting device, and connects various parts of the entire detecting device using various interfaces and lines.
  • the memory can be used to store the computer program and/or module, the processor implementing the detection by running or executing a computer program and/or module stored in the memory, and recalling data stored in the memory Various functions of the device.
  • the memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application required for at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may be stored. Data created based on the use of the terminal (such as audio data, phone book, etc.).
  • the memory may include a high-speed random access memory, and may also include a non-volatile memory such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SMC), and a Secure Digital (SD) card. Flash Card, at least one disk storage device, flash device, or other volatile solid-state storage device.
  • a non-volatile memory such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SMC), and a Secure Digital (SD) card.
  • Flash Card at least one disk storage device, flash device, or other volatile solid-state storage device.
  • An embodiment of the present invention further provides an aircraft, wherein the aircraft may include:
  • a power system disposed on the fuselage for providing flight power to the drone
  • a detecting device according to any of the above.
  • the power system of the aircraft may include: a motor, an ESC, a propeller, etc.
  • the aircraftless aircraft may further include a payload, such as an imaging device, an infrared imager, etc., wherein the payload may be coupled to the aircraft via a carrier, wherein the carrier Can be a gimbal.
  • the moving object described in the above method embodiment is an aircraft, and the detecting device is disposed on the aircraft.
  • the aircraft can realize fast and accurate detection of the mirror object, so as to mirror the object during flight. Perform obstacle avoidance operations, or perform task execution on mirrored objects, or perform safe landings.
  • the present invention also provides a computer readable storage medium having stored thereon a computer program.
  • the processor can be configured to perform the following steps:
  • the polarization component is disposed on the image acquisition component, and the polarization direction of the polarization component when acquiring the first image is different from the polarization direction of the polarization component when acquiring the first image;
  • the image acquisition assembly includes two image sensing devices, wherein each of the image sensing devices is provided with a polarizing element;
  • the processor may be specifically configured to perform the following steps:
  • a second image is acquired by a second one of the two image sensing devices.
  • the processor when the computer program stored in the computer readable storage medium is executed by the processor, the processor may be specifically configured to perform the following steps:
  • a computer program stored in a computer readable storage medium is processed
  • the processor can be specifically configured to perform the following steps:
  • a specular object is determined within the field of view of the image acquisition component based on the third image.
  • the processor when the computer program stored in the computer readable storage medium is executed by the processor, the processor may be specifically configured to perform the following steps:
  • the pixel value of each pixel in the first image and the pixel value of the corresponding pixel in the second image are subtracted, and the image obtained after the subtraction process is determined as the third image.
  • the processor when the computer program stored in the computer readable storage medium is executed by the processor, the processor may be specifically configured to perform the following steps:
  • the pixel value of each pixel in the first image is divided into the pixel value of the corresponding pixel in the second image, and the image obtained after the division processing is determined as the third image.
  • the processor when the computer program stored in the computer readable storage medium is executed by the processor, the processor may be specifically configured to perform the following steps:
  • a correlation between the third image and the preset image is detected, and whether a specular object exists within the field of view of the image capturing component is determined according to the correlation.
  • the processor when the computer program stored in the computer readable storage medium is executed by the processor, the processor may be specifically configured to perform the following steps:
  • the processor when the computer program stored in the computer readable storage medium is executed by the processor, the processor may be specifically configured to perform the following steps:
  • the functionality of the detection device if implemented in the form of a software functional unit and sold or used as a standalone product, may be stored in a computer readable storage medium.
  • the present invention implements all or part of the processes in the foregoing embodiments, and may also be completed by a computer program to instruct related hardware.
  • the computer program may be stored in a computer readable storage medium.
  • the steps of the various method embodiments described above may be implemented when the program is executed by the processor.
  • the computer program comprises computer program code, which may be in the form of source code, object code form, executable file or some intermediate form.
  • the computer can The reading medium may include any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a removable hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), and a random access. Memory (RAM, Random Access Memory), electrical carrier signals, telecommunications signals, and software distribution media. It should be noted that the content contained in the computer readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in a jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, computer readable media Does not include electrical carrier signals and telecommunication signals.
  • 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 in an electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present 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.

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Optics & Photonics (AREA)
  • Quality & Reliability (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)

Abstract

一种检测方法，包括：通过图像采集组件采集第一图像和第二图像，其中，图像采集组件上设置偏振元件，获取第一图像时偏振元件的偏振方向不同于获取第二图像时偏振元件的偏振方向；根据第一图像、第二图像，确定图像采集组件的视场范围内是否存在镜面物体。还涉及一种检测设备以及一种飞行器。该检测方法可简便、快捷、准确地检测出水面等镜面物体。

Description

一种检测方法、检测设备以及飞行器 技术领域
本申请实施例涉及无人机技术领域，尤其涉及一种检测方法、检测设备以及飞行器。
背景技术
在无人机的飞行环境中，往往会存在镜面物体，镜面物体对无人机的飞行安全带来较大的安全隐患。例如，无人机目前主要使用视觉传感器来采集障碍物的图像实现避障，然而，使用视觉传感器不能很好地采集到镜面物体的图像，因为视觉传感器采集到的图像往往是由镜面物体反射的光形成的图像，因此不能对镜面物体进行准确分析，容易产生误判。目前缺乏对镜面物体的检测方法，以避免由镜面物体可能引起的无人机的安全飞行事故。
发明内容
本发明实施例提供了一种检测方法、检测设备以及飞行器，用于简便、快捷、准确地检测出镜面物体。
有鉴于此，本发明第一方面提供一种检测方法，可包括：
通过图像采集组件采集第一图像和第二图像，其中，图像采集组件上设置偏振元件，获取第一图像时偏振元件的偏振方向不同于获取第二图像时偏振元件的偏振方向；
根据第一图像、第二图像，确定图像采集组件的视场范围内是否存在镜面物体。
本发明第二方面提供一种检测设备，可包括：
图像采集组件和处理器；
其中，图像采集组件，用于：
采集第一图像和第二图像，其中，图像采集组件上设置偏振元件，获取第一图像时偏振元件的偏振方向不同于获取第二图像时偏振元件的偏振方向；
处理器，用于：
根据第一图像、第二图像，确定图像采集组件的视场范围内是否存在镜面 物体。
本发明第三方面提供一种飞行器，可包括：
机身；
设置在机身上的动力系统，用于为飞行器提供飞行动力；
如第二方面的检测设备。
从以上技术方案可以看出，本发明实施例具有以下优点：
区别于现有技术的情况，本发明实施例中，通过利用图像采集组件上的偏振元件，可以在不同偏振方向获取第一图像和第二图像，而后可以根据第一图像、第二图像，确定图像采集组件的视场范围内是否存在镜面物体。获取第一图像时偏振元件的偏振方向不同于获取第一图像时偏振元件的偏振方向，当图像采集组件的视场范围内存在镜面物体时，获取第一图像时进入图像采集组件的传入光线不同于获取第二图像时进入图像采集组件的传入光线，使得第一图像和第二图像之间的存在明显的差异，这样可以根据第一图像与第二图像简便、快捷、准确地检测出图像采集组件的视场范围内是否存在镜面物体。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例描述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为本发明实施例中检测方法一个实施例示意图；
图2为本发明实施例中第一图像的光强分布示意图；
图3为本发明实施例中第二图像的光强分布示意图；
图4为本发明实施例中检测方法另一实施例示意图；
图5为本发明实施例中第一图像与第二图像的第一预设处理示意图；
图6为本发明实施例中第一图像与第二图像的第二预设处理示意图；
图7为本发明实施例中检测方法另一实施例示意图；
图8为本发明实施例中检测设备一个实施例示意图。
具体实施方式
本发明实施例提供了一种检测方法、检测设备以及飞行器，用于简便、快捷、准确地检测出镜面物体。
为了使本技术领域的人员更好地理解本发明方案，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，所描述的实施例仅仅是本发明一部分的实施例，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做出创造性劳动前提下所获得的所有其他实施例，都应当属于本发明保护的范围。
本发明的说明书和权利要求书及上述附图中的术语“第一”、“第二”、“第三”、“第四”等(如果存在)是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换，以便这里描述的实施例能够以除了在这里图示或描述的内容以外的顺序实施。此外，术语“包括”和“具有”以及他们的任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
为便于理解，下面对本发明实施例中的具体流程进行描述，请参阅图1，本发明实施例中检测方法一个实施例包括：
101、通过图像采集组件采集第一图像和第二图像；
本实施例中，所述方法可以由检测设备来执行，其中检测设备包括图像采集组件，图像采集组件上设置有偏振元件，在图像采集组件进行图像采集时，可以利用偏振元件的不同偏振方向分别进行第一图像和第二图像的采集，即获取第一图像时偏振元件的偏振方向不同于获取第二图像时偏振元件的偏振方向。
具体的，当自然光(全向的)斜射到镜面物体发生镜面反射时，反射光会发生偏振效果，即一般为部分偏振光(特殊情况下为线偏振光)。相对于自然光的轴对称分布的光强，部分偏振光的振动方向相对于传播方向具有不对称性，部分偏振光光强的分布则为非轴对称。偏振元件上设有一个特殊的方向，即偏振方向，只有振动方向与偏振方向平行的光才能通过偏振元件，因而，由于获取第一图像和第二图像时偏振元件的偏振方向不同，则第一图像和第二图像将具有不同的特征，如光强分布不同。
可以理解的是，图像采集组件采集第一图像和第二图像的光线来源应一致，同时，第一图像和第二图像的图像内容至少存在交叉部分，以能够结合第一图像和第二图像对图像采集组件的视场范围内的物体进行正确而有效的检测。优选的，第一图像的图像内容可以和第二图像的图像内容完全重合，即第一图像和第二图像除通过不同偏振方向获取之外，其它条件相同。
进一步的，本实施例中，获取图像采集组件采集的第一图像和第二图像的具体方式可以为：
第一种方式：
通过两个图像感测器件中的第一器件采集第一图像；
通过两个图像感测器件中的第二器件采集第二图像。
具体的，图像采集组件可以包括两个图像感测器件，其中，每一个图像感测器件上配置一个偏振元件。该图像感测器件可以包括光学镜头，即图像采集组件可以包括两个光学镜头，如双目摄像头，两个光学镜头的表面上可以分别覆盖一层诸如偏振片(偏振元件)，两层偏振片的起偏角不同，这样可以使得两层偏振片的偏振方向不一致，从而在两个光学镜头进行摄制时，进入两个光学镜头的光线的偏振方向会不相同，进而实现在偏振片的不同偏振方向分别采集第一图像和第二图像。
可以理解的是，在实际应用中，本实施例中的图像采集组件还可以包括两个以上的图像感测器件，其中，至少两个图像感测器件中的每一个图像感测器件上可以配置有一个偏振元件，每一个偏振元件的偏振方向不同，从而可以保证在至少两个偏振元件的两个偏振方向获取到第一图像和第二图像，所使用的配置有偏振元件的图像感测器件越多，图像采集组件的视场范围内的物体的面积越大，则精度越高，具体此处不做限定。
第二种方式：
将偏振元件的偏振方向设置为第一偏振方向，通过图像采集组件采集第一图像；
将偏振元件的偏振方向设置为第二偏振方向，通过图像采集组件采集第二图像；
具体的，图像采集组件可以为诸如单目摄像头，该单目摄像头可以配置有一个偏振元件，该偏振元件的起偏角可以进行调整，从而可以调整为不同的偏 振方向。在实际应用中。可以先将该偏振元件的偏振方向设置为第一偏振方向，并在该第一偏振方向下通过图像采集组件采集第一图像，而后可以将偏振元件的偏振方向从第一偏振方向调整为第二偏振方向，并在该第二偏振方向下通过图像采集组件采集第二图像。
进一步的，在该种获取方式下，偏振元件可以为液晶装置，液晶装置是一种能通过电信号控制偏振方向的装置。因此，在本实施例中，可以加以利用液晶装置的通电情况进行偏振方向的切换，如偏振横向获取第一图像后，可以切换为偏振纵向并获取第二图像，则可以得到在液晶装置的两个不同偏振方向获取的图像。
可以理解的是，在偏振元件为一个时，本实施例中的偏振元件除了上述说明的液晶装置，在实际应用中，还可以为其它类型的偏振元件，只要能够进行偏振方向的切换即可，具体此处不做限定。
进一步的，本实施例中，图像采集组件也可以包括一个以上的图像感测器件，其中至少一个图像感测器件上配置有一个可以切换偏振方向的偏振元件，如液晶装置，所使用的配置有偏振元件的图像感测器件越多，图像采集组件的视场范围内的物体的面积越大，则精度越高。
需要说明的是，本实施例中获取图像采集组件采集的第一图像和第二图像的具体方式除了上述说明的几个例子，在实际应用中，还可以采用其它方式，只要获取第一图像时偏振元件的偏振方向不同于获取第一图像时偏振元件的偏振方向即可，具体此处不做限定。
本实施例中，图像采集组件所在的检测设备可以配置于移动物体上，如飞行器、遥控汽车等移动物体，从而可以较好地协助飞行器、遥控汽车等避开镜面物体，完成避障操作，以防止安全事故的发生。
102、根据第一图像、第二图像，确定图像采集组件的视场范围内是否存在镜面物体。
本实施例中，通过图像采集组件采集第一图像后，可以根据第一图像、第二图像，确定图像采集组件的视场范围内是否存在镜面物体。
具体的，当自然光射向镜面物体时，会发生反射和折射，光的偏振状态会改变。通常情况下，反射光和折射光不再是自然光，而是部分偏振光，而且在反射光中垂直于入射面的光振动要多于平行振动，而折射光则相反。反射光的 偏振化程度与入射角有关，当入射角度等于布儒斯特角时，反射光就成为只有垂直于入射面的线偏振光。当获取第一图像时偏振元件的偏振方向不同于获取第一图像时偏振元件的偏振方向时，第一图像和第二图像会有明显的差异。例如，假设图像采集组件的视场范围内存在镜面物体，图像采集组件上配置有两个图像感测器件，即为双目摄像头，在该双目摄像头上，该双目摄像头的每一个光学摄像头上可以分别配置有起偏角互相垂直的偏振片，当镜面物体反射的光线传入双目摄像头时，双目摄像头可以利用透过偏振片的两个偏振方向的光线获取第一图像和第二图像。如图2和图3所示，假设双目摄像头的视场范围内存在镜面物体，如镜面，那么在自然光的情况下，可以得到在一个偏振片的第一偏振方向获取的第一图像(如图2)，以及在另一个偏振片的第二偏振方向获取的第二图像(如图3)，第一图像和第二图像的光强分布特征不一致。反之，若图像采集组件的视场范围内不存在镜面物体，则第一图像和第二图像之间的光强分布特征将一致。由此可知，根据获取的第一图像和第二图像，可以确定图像采集组件的视场范围内是否存在镜面物体。
本实施例中，镜面物体可以包括水面、玻璃、镜子、瓷砖中的至少一种。
本发明实施例中，通过利用图像采集组件上的偏振元件，可以在不同偏振方向获取第一图像和第二图像，而后可以根据第一图像、第二图像，确定图像采集组件的视场范围内是否存在镜面物体。获取第一图像时偏振元件的偏振方向不同于获取第一图像时偏振元件的偏振方向，当图像采集组件的视场范围内存在镜面物体时，获取第一图像时进入图像采集组件的传入光线不同于获取第二图像时进入图像采集组件的传入光线，使得第一图像和第二图像之间的存在明显的差异，这样可以根据第一图像与第二图像简便、快捷、准确地检测出图像采集组件的视场范围内是否存在镜面物体。
可以理解的是，基于对第一图像、第二图像的不同处理方式，确定图像采集组件的视场范围内是否存在镜面物体的方式可以不同，下面进行具体说明：
请参阅图4，本发明实施中检测方法另一实施例包括：
401、通过图像采集组件采集第一图像和第二图像；
本实施例中的步骤401与图1所示实施例中的步骤101相同，此处不再赘述。
402、对第一图像和第二图像进行预设处理以获取第三图像；
本实施例中，获取图像采集组件采集的第一图像和第二图像后，可以对第一图像和第二图像进行预设处理以获取第三图像。
本实施例中，对第一图像和第二图像进行预设处理以获取第三图像的具体方式可以为：
将第一图像中每一个像素点的像素值与第二图像中对应的像素点的像素值作减法处理，将减法处理后获取的图像确定为第三图像；或，
将第一图像中每一个像素点的像素值与第二图像中对应的像素点的像素值作除法处理，将除法处理后获取的图像确定为第三图像。
具体的，一张图像可以包含多个像素点，每一个像素点的值可以反映该点图像的亮度。那么通过分别确定第一图像和第二图像中每一个像素点的像素值，可以分别确定第一图像和第二图像中相应位置的亮度。为了准确判断第一图像和第二图像显示的光强分布之间的差异，可以将第一图像中每一个像素点的像素值与第二图像中对应的像素点的像素值作减法处理，以减法处理后获取的图像为第三图像，或者，也可以将第一图像中每一个像素点的像素值与第二图像中对应的像素点的像素值作除法处理，以除法处理后获取的图像为第三图像。通过获取到的第三图像，可以为准确地显示第一图像中像素点的像素值和第二图像中对应的像素点的像素值之间的差异。
可以理解的是，本实施例中对第一图像和第二图像进行预设处理以获取第三图像的具体方式除了上述说明的几个例子，在实际应用中，还可以采用其它算法，只要使得获取的第三图像能够用于确定图像采集组件的视场范围内是否存在镜面物体即可，具体此处不做限定。
本实施例中像素点的像素值的计算方式可以参照现有技术，此处不再赘述。
403、根据第三图像确定图像采集组件的视场范围内是否存在镜面物体。
本实施例中，对第一图像和第二图像进行预设处理以获取第三图像后，可以根据第三图像确定图像采集组件的视场范围内是否存在镜面物体。
具体的，若第一图像和第二图像的光强分布特征一致，则对第一图像和第二图像进行预设处理以获取第三图像后，第三图像由于光强分布特征引起的图案分布将会整体发生变化，如第三图像相对第一图像或第二图像而言，第三图 像将整体变暗。反之，若第一图像和第二图像的光强分布特征不一致，则对第一图像和第二图像进行预设处理以获取第三图像后，第三图像由于光强分布特征引起的图案分布不仅明暗度会发生变化，形状也会发生变化，如沿用图1所示实施例中步骤102中说明的第一图像(图2)和第二图像(图3)，当图像采集组件的视场范围内存在镜面物体时，若将第一图像中每一个像素点的像素值与第二图像中对应的像素点的像素值作减法处理，可以得到如图5所示的第三图像，若将第一图像中每一个像素点的像素值与第二图像中对应的像素点的像素值作除法处理，可以得到如图6所示的第三图像。由此可见，当图像采集组件的视场范围内存在镜面物体时，第三图像将呈现出不同于第一图像、第二图像的显示特征。
进一步的，本实施例中，为了提高镜面物体的识别准确率，根据第三图像确定图像采集组件的视场范围内是否存在镜面物体的具体方式可以为：
检测第三图像与预设图像之间的相关性，根据相关性确定图像采集组件的视场范围内是否存在镜面物体；或，
将第三图像输入预设的第一运算模型，根据第一运算模型输出的识别结果确定图像采集组件的视场范围内是否存在镜面物体。
具体的，针对第一种确定方式，这里以水面作为镜面物体来进行示意性说明，在确定第三图像与预设图像的相关性之前，需要确定预设图像，其中具体过程可以为：使用如前所述的图像采集组件对水面进行拍摄，获取多张包含水面的第一标准图像A1、A2、A3、A4、A5，其中，第一标准图像对应的偏振元件的偏振方向为第一偏振方向，获取多张包含水面的第二标准图像B1、B2、B3、B4、B5，其中，第二标准图像对应的偏振元件的偏振方向为第二偏振方向。进一步地，可以将A1与B1、A2与B2、A3与B3、A4与B4、A5与B5作预设处理，如对相应像素点的像素值作减法处理或除法处理，以得到5张预设处理后的图像，再将这5张预设处理后的图像进行平均处理，可以得到用于检测图像采集组件的视场范围内是否存在水面的预设图像。因而，本实施例中，获取到第三图像后，可以确定第三图像与水面的预设图像的相关性，当第三图像与水面的预设图像的相关性大于预设阈值时，可以确定图像采集组件的视场范围内存在水面，反之，则可以确定图像采集组件的视场范围内不存在水面。
需要说明的是，本实施例中水面的第一标准图像、第二标准图像的张数还 可以为其它数值，上述仅为举例说明，而预设图像的获取方式除了采用上述说明的内容，在实际应用中，还可以采用其它方式，只要能够得到用于检测第三图像的相关性的预设图像即可，具体此处不做限定。
可以理解的是，在实际应用中，不同镜面物体的预设图像可以分类存储，同时，在检测第三图像预设图像的相关性时，针对不同的镜面物体，可以分别确定第三图像与相应预设图像的相关性，如若有水面、玻璃、镜面的预设图像，则可以得到三个相关性，并可以根据这三个相关性分别确定图像采集组件的视场范围内是否存在镜面物体，以在确定存在镜面物体时，还能够确定镜面物体为何物。
基于另一种确定方式，本实施例中，需要首先对第一运算模型进行训练，其中第一运算模型可以为神经网络、分类器等。沿用上述说明的例子，将A1与B1、A2与B2、A3与B3、A4与B4、A5与B5作预设处理，如对相应像素点的像素值作减法处理或除法处理，以得到5张预设处理后的图像，可以将这5张预设处理后的图像作为训练样本，对第一运算模型进行训练，当完成第一运算模型的训练后，即可以使用训练后的第一运算模型对第三图像进行检测，训练后的第一运算模型可以输出识别结果，例如识别结果可以为置信度，当置信度高于预设的阈值时，可以判定图像采集组件的视场范围内存在水面。
需要说明的是，本实施例中训练样本除了包括预设处理后的水面的标准图像，还可以包括预设处理后的其它镜面物体的标准图像，更可以包括预设处理后的非镜面物体的标准图像，以有利于区别不同镜面物体以及镜面物体与非镜面物体的特征信息，如瓷砖的规则的条纹信息，从而使得训练后的第一运算模型可以根据输入的第三图像进行镜面物体的准确判断，并可以进一步识别镜面物体为何物。
可以理解的是，本实施例中根据第三图像确定图像采集组件的视场范围内是否存在镜面物体的具体方式除了上述说明的几个例子，在实际应用中，还可以采用其它方式进行结合或单独使用，只要能够根据第三图像确定图像采集组件的视场范围内是否存在镜面物体即可，具体此处不做限定。
请参阅图7，本发明实施中检测方法另一实施例包括：
701、通过图像采集组件采集第一图像和第二图像；
本实施例中的步骤701与图1所示实施例中的步骤101相同，此处不再赘 述。
702、将第一图像和第二图像输入预设的第二运算模型，根据第二运算模型输出的识别结果确定图像采集组件的视场范围内是否存在镜面物体。
本实施例中，通过图像采集组件采集第一图像和第二图像后，可以将第一图像和第二图像输入预设的第二运算模型，根据第二运算模型输出的识别结果确定图像采集组件的视场范围内是否存在镜面物体。
具体的，为了准确判断镜面物体，需要首先对第二运算模型进行训练，其中第二运算模型可以为神经网络、分类器等。这里以水面作为镜面物体来进行示意性说明，在训练时，使用如前所述的图像采集组件对水面进行拍摄，获取多张包含水面的第一标准图像，其中，第一标准图像对应的偏振元件的偏振方向为第一偏振方向，获取多张包含水面的第二标准图像，其中，第一标准图像对应的偏振元件的偏振方向为第二偏振方向，使用第一标准图像和第二标准图像作为训练样本，对第二运算模型进行训练。当完成第二运算模型的训练后，即可以使用训练后的第二运算模型对第一图像和第二图像进行检测，第二运算模型可以输出识别结果，例如识别结果可以为置信度，当置信度高于预设的阈值时，可以判定图像采集组件的视场范围内存在水面。
需要说明的是，本实施例中训练样本除了包括水面的标准图像，还可以包括其它镜面物体的标准图像，更可以包括非镜面物体的标准图像，以有利于区别不同镜面物体以及镜面物体与非镜面物体的特征信息，如瓷砖的规则的条纹信息，从而使得训练后的第二运算模型可以根据输入的第一图像、第二图像进行镜面物体的准确判断，并可以进一步识别镜面物体为何物。
需要说明的是，本发明仅以图4所示实施例和图7所示实施例对镜面物体的检测方法进行了举例说明，在实际应用中，还可以采用其它方法，只要能够根据第一图像和第二图像确定图像采集组件的视场范围内是否存在镜面物体即可，具体此处不做限定。
上面对本发明实施例中的检测方法进行了描述，下面从硬件处理的角度对本发明实施例中的检测设备分别进行描述，请参阅图8，本发明实施例中的检测设备一个实施例包括：
图像采集组件801和处理器802；
其中，图像采集组件801，用于：
采集第一图像和第二图像，其中，图像采集组件上设置偏振元件，获取第一图像时偏振元件的偏振方向不同于获取第一图像时偏振元件的偏振方向；
处理器802，用于：
根据第一图像、第二图像，确定图像采集组件的视场范围内是否存在镜面物体。
可选的，在本发明的一些实施例中，图像采集组件包括两个图像感测器件，其中，每一个图像感测器件上配置一个偏振元件；
图像采集组件801，可以进一步具体用于：
通过两个图像感测器件中的第一器件采集第一图像；
通过两个图像感测器件中的第二器件采集第二图像。
可选的，在本发明的一些实施例中，处理器802，可以进一步具体用于：
将偏振元件的偏振方向设置为第一偏振方向；
图像采集组件801，可以进一步具体用于：采集第一图像；
处理器802，可以进一步具体用于：
将偏振元件的偏振方向设置为第二偏振方向；
图像采集组件801，可以进一步具体用于：采集第二图像。
可选的，在本发明的一些实施例中，处理器802，可以进一步具体用于：
对第一图像和第二图像进行预设处理以获取第三图像；
根据第三图像确定图像采集组件的视场范围内是否存在镜面物体。
可选的，在本发明的一些实施例中，处理器802，可以进一步具体用于：
将第一图像中每一个像素点的像素值与第二图像中对应的像素点的像素值作减法处理，将减法处理后获取的图像确定为第三图像。
可选的，在本发明的一些实施例中，处理器802，可以进一步具体用于：
将第一图像中每一个像素点的像素值与第二图像中对应的像素点的像素值作除法处理，将除法处理后获取的图像确定为第三图像。
可选的，在本发明的一些实施例中，处理器802，可以进一步具体用于：
检测第三图像与预设图像之间的相关性，根据相关性确定图像采集组件的视场范围内是否存在镜面物体。
可选的，在本发明的一些实施例中，处理器802，可以进一步具体用于：
将第三图像输入预设的第一运算模型，根据第一运算模型输出的识别结果 确定图像采集组件的视场范围内是否存在镜面物体。
可选的，在本发明的一些实施例中，处理器802，可以进一步具体用于：
将第一图像和第二图像输入预设的第二运算模型，根据第二运算模型输出的识别结果确定图像采集组件的视场范围内是否存在镜面物体。
进一步的，本实施例中还可以包括存储器803，该存储器803上可以存储有计算机程序，以使得处理器802用于执行存储器803中存储的计算机程序时，可以实现上述说明的步骤。
其中，所述检测设备可包括，但不仅限于，图像采集组件、处理器、存储器。本领域技术人员可以理解，图8示意图仅仅是检测设备的示例，并不构成对检测设备的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件，例如所述检测设备还可以包括输入输出设备、网络接入设备等。
所述处理器可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable GateArray，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等，所述处理器是所述检测设备的控制中心，利用各种接口和线路连接整个检测设备的各个部分。
所述存储器可用于存储所述计算机程序和/或模块，所述处理器通过运行或执行存储在所述存储器内的计算机程序和/或模块，以及调用存储在存储器内的数据，实现所述检测设备的各种功能。所述存储器可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序(比如声音播放功能、图像播放功能等)等；存储数据区可存储根据终端的使用所创建的数据(比如音频数据、电话本等)等。此外，存储器可以包括高速随机存取存储器，还可以包括非易失性存储器，例如硬盘、内存、插接式硬盘，智能存储卡(SmartMedia Card,SMC)，安全数字(Secure Digital,SD)卡，闪存卡(Flash Card)、至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。
本发明实施例还提供一种飞行器，其中飞行器可以包括：
机身
设置在机身上的动力系统，用于为无人机提供飞行动力；
如上所述的任一项的检测设备。
具体地，飞行器的动力系统可以包括：电机、电调、螺旋桨等，其中无飞行器还可以包括有效负载，例如成像装置、红外成像仪等，其中有效负载可以通过承载件与飞行器连接，其中承载件可以为云台。
即上述方法实施例中说明的移动物体为飞行器，飞行器上配置有检测设备，通过检测设备中的图像采集组件，飞行器可以实现对镜面物体的快捷、准确的检测，以在飞行过程中对镜面物体进行避障操作，或实现对镜面物体的任务执行，或进行安全降落等。
本发明还提供了一种计算机可读存储介质，其上存储有计算机程序，计算机程序被处理器执行时，处理器，可以用于执行如下步骤：
通过图像采集组件采集第一图像和第二图像，其中，图像采集组件上设置偏振元件，获取第一图像时偏振元件的偏振方向不同于获取第一图像时偏振元件的偏振方向；
根据第一图像、第二图像，确定图像采集组件的视场范围内是否存在镜面物体。
在本发明的一些实施例中，图像采集组件包括两个图像感测器件，其中，每一个图像感测器件上配置一个偏振元件；
计算机可读存储介质存储的计算机程序被处理器执行时，处理器，可以具体用于执行如下步骤：
通过两个图像感测器件中的第一器件采集第一图像；
通过两个图像感测器件中的第二器件采集第二图像。
在本发明的一些实施例中，计算机可读存储介质存储的计算机程序被处理器执行时，处理器，可以具体用于执行如下步骤：
将偏振元件的偏振方向设置为第一偏振方向，通过图像采集组件采集第一图像；
将偏振元件的偏振方向设置为第二偏振方向，通过图像采集组件采集第二图像；
在本发明的一些实施例中，计算机可读存储介质存储的计算机程序被处理 器执行时，处理器，可以具体用于执行如下步骤：
对第一图像和第二图像进行预设处理以获取第三图像；
根据第三图像确定图像采集组件的视场范围内是否存在镜面物体。
在本发明的一些实施例中，计算机可读存储介质存储的计算机程序被处理器执行时，处理器，可以具体用于执行如下步骤：
将第一图像中每一个像素点的像素值与第二图像中对应的像素点的像素值作减法处理，将减法处理后获取的图像确定为第三图像。
在本发明的一些实施例中，计算机可读存储介质存储的计算机程序被处理器执行时，处理器，可以具体用于执行如下步骤：
将第一图像中每一个像素点的像素值与第二图像中对应的像素点的像素值作除法处理，将除法处理后获取的图像确定为第三图像。
在本发明的一些实施例中，计算机可读存储介质存储的计算机程序被处理器执行时，处理器，可以具体用于执行如下步骤：
检测第三图像与预设图像之间的相关性，根据相关性确定图像采集组件的视场范围内是否存在镜面物体。
在本发明的一些实施例中，计算机可读存储介质存储的计算机程序被处理器执行时，处理器，可以具体用于执行如下步骤：
将第三图像输入预设的第一运算模型，根据第一运算模型输出的识别结果确定图像采集组件的视场范围内是否存在镜面物体。
在本发明的一些实施例中，计算机可读存储介质存储的计算机程序被处理器执行时，处理器，可以具体用于执行如下步骤：
将第一图像和第二图像输入预设的第二运算模型，根据第二运算模型输出的识别结果确定图像采集组件的视场范围内是否存在镜面物体。
可以理解的是，所述检测设备的功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明实现上述实施例方法中的全部或部分流程，也可以通过计算机程序来指令相关的硬件来完成，所述的计算机程序可存储于一计算机可读存储介质中，该计算机程序在被处理器执行时，可实现上述各个方法实施例的步骤。其中，所述计算机程序包括计算机程序代码，所述计算机程序代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可 读介质可以包括：能够携带所述计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，RandomAccess Memory)、电载波信号、电信信号以及软件分发介质等。需要说明的是，所述计算机可读介质包含的内容可以根据司法管辖区内立法和专利实践的要求进行适当的增减，例如在某些司法管辖区，根据立法和专利实践，计算机可读介质不包括电载波信号和电信信号。
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。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统，装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本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。
以上所述，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 (23)

  1. 一种检测方法，其特征在于，包括：
    通过图像采集组件采集第一图像和第二图像，其中，所述图像采集组件上设置偏振元件，获取第一图像时偏振元件的偏振方向不同于获取第二图像时偏振元件的偏振方向；
    根据所述第一图像、所述第二图像，确定所述图像采集组件的视场范围内是否存在镜面物体。
  2. 根据权利要求1所述的方法，其特征在于，所述图像采集组件包括两个图像感测器件，其中，每一个图像感测器件上配置一个偏振元件；
    所述通过图像采集组件采集第一图像和第二图像包括：
    通过所述两个图像感测器件中的第一器件采集第一图像；
    通过所述两个图像感测器件中的第二器件采集第二图像。
  3. 根据权利要求1所述的方法，其特征在于，所述通过图像采集组件采集第一图像和第二图像包括：
    将所述偏振元件的偏振方向设置为第一偏振方向；
    通过图像采集组件采集第一图像；
    将所述偏振元件的偏振方向设置为第二偏振方向；
    通过所述图像采集组件采集第二图像；
  4. 根据权利要求3所述的方法，其特征在于，所述偏振元件为液晶装置。
  5. 根据权利要求1至4中任一项所述的方法，其特征在于，所述根据所述第一图像、所述第二图像，确定所述图像采集组件的视场范围内是否存在镜面物体包括：
    对所述第一图像和所述第二图像进行预设处理以获取第三图像；
    根据所述第三图像确定所述图像采集组件的视场范围内是否存在镜面物体。
  6. 根据权利要求5所述的方法，其特征在于，所述对所述第一图像和所述第二图像进行预设处理以获取第三图像包括：
    将所述第一图像中每一个像素点的像素值与所述第二图像中对应的像素 点的像素值作减法处理，将所述减法处理后获取的图像确定为第三图像。
  7. 根据权利要求6所述的方法，其特征在于，所述对所述第一图像和所述第二图像进行预设处理以获取第三图像包括：
    将所述第一图像中每一个像素点的像素值与所述第二图像中对应的像素点的像素值作除法处理，将所述除法处理后获取的图像确定为第三图像。
  8. 根据权利要求5至7中任一项所述的方法，其特征在于，所述根据所述第三图像确定所述图像采集组件的视场范围内是否存在镜面物体包括：
    检测所述第三图像与预设图像之间的相关性，根据所述相关性确定所述图像采集组件的视场范围内是否存在镜面物体。
  9. 根据权利要求5至7中任一项所述的方法，其特征在于，所述根据所述第三图像确定所述图像采集组件的视场范围内是否存在镜面物体包括：
    将所述第三图像输入预设的第一运算模型，根据所述第一运算模型输出的识别结果确定所述图像采集组件的视场范围内是否存在镜面物体。
  10. 根据权利要求1至9中任一项所述的物体检测方法，其特征在于，所述根据所述第一图像、所述第二图像，判断所述图像采集组件的视场范围内存在镜面物体包括：
    将所述第一图像和所述第二图像输入预设的第二运算模型，根据所述第二运算模型输出的识别结果确定所述图像采集组件的视场范围内是否存在镜面物体。
  11. 根据权利要求1至10中任一项所述的检测方法，其特征在于，所述镜面物体包括水面、玻璃、镜子、瓷砖中的至少一种。
  12. 一种检测设备，其特征在于，包括：
    图像采集组件和处理器；
    其中，所述图像采集组件，用于：
    采集第一图像和第二图像，其中，所述图像采集组件上设置偏振元件，获取第一图像时偏振元件的偏振方向不同于获取第二图像时偏振元件的偏振方向；
    所述处理器，用于：
    根据所述第一图像、所述第二图像，确定所述图像采集组件的视场范围内是否存在镜面物体。
  13. 根据权利要求12所述的装置，其特征在于，所述图像采集组件包括两个图像感测器件，其中，每一个图像感测器件上配置一个偏振元件；
    所述图像采集组件，具体用于：
    通过所述两个图像感测器件中的第一器件采集第一图像；
    通过所述两个图像感测器件中的第二器件采集第二图像。
  14. 根据权利要求12所述的装置，其特征在于，所述处理器，具体用于：将所述偏振元件的偏振方向设置为第一偏振方向；
    所述图像采集组件，具体用于：采集第一图像；
    所述处理器，具体用于：将所述偏振元件的偏振方向设置为第二偏振方向；
    所述图像采集组件，具体用于：采集第二图像；
  15. 根据权利要求14所述的方法，其特征在于，所述偏振元件为液晶装置。
  16. 根据权利要求12至15中任一项所述的装置，其特征在于，所述处理器，具体用于：
    对所述第一图像和所述第二图像进行预设处理以获取第三图像；
    根据所述第三图像确定所述图像采集组件的视场范围内是否存在镜面物体。
  17. 根据权利要求16所述的装置，其特征在于，所述处理器，具体用于：
    将所述第一图像中每一个像素点的像素值与所述第二图像中对应的像素点的像素值作减法处理，将所述减法处理后获取的图像确定为第三图像。
  18. 根据权利要求16所述的装置，其特征在于，所述处理器，具体用于：
    将所述第一图像中每一个像素点的像素值与所述第二图像中对应的像素点的像素值作除法处理，将所述除法处理后获取的图像确定为第三图像。
  19. 根据权利要求16至18中任一项所述的装置，其特征在于，所述处理器，具体用于：
    检测所述第三图像与预设图像之间的相关性，根据所述相关性确定所述图像采集组件的视场范围内是否存在镜面物体。
  20. 根据权利要求16至18中任一项所述的装置，其特征在于，所述处理器，具体用于：
    将所述第三图像输入预设的第一运算模型，根据所述第一运算模型输出的 识别结果确定所述图像采集组件的视场范围内是否存在镜面物体。
  21. 根据权利要求12至20中任一项所述的装置，其特征在于，所述处理器，具体用于：
    将所述第一图像和所述第二图像输入预设的第二运算模型，根据所述第二运算模型输出的识别结果确定所述图像采集组件的视场范围内是否存在镜面物体。
  22. 根据权利要求12至21中任一项所述的装置，其特征在于，所述镜面物体包括水面、玻璃、镜子、瓷砖中的至少一种。
  23. 一种飞行器，其特征在于，包括：
    机身；
    设置在所述机身上的动力系统，用于为飞行器提供飞行动力；
    如权利要求12至22中任一项所述的检测设备。
PCT/CN2017/090778 2017-06-29 2017-06-29 一种检测方法、检测设备以及飞行器 WO2019000313A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201780005493.XA CN108701356A (zh) 2017-06-29 2017-06-29 一种检测方法、检测设备以及飞行器
PCT/CN2017/090778 WO2019000313A1 (zh) 2017-06-29 2017-06-29 一种检测方法、检测设备以及飞行器

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/090778 WO2019000313A1 (zh) 2017-06-29 2017-06-29 一种检测方法、检测设备以及飞行器

Publications (1)

Publication Number Publication Date
WO2019000313A1 true WO2019000313A1 (zh) 2019-01-03

Family

ID=63844068

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/090778 WO2019000313A1 (zh) 2017-06-29 2017-06-29 一种检测方法、检测设备以及飞行器

Country Status (2)

Country Link
CN (1) CN108701356A (zh)
WO (1) WO2019000313A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006058122A (ja) * 2004-08-19 2006-03-02 Nagoya Electric Works Co Ltd 路面状態判別方法およびその装置
CN101033961A (zh) * 2007-02-06 2007-09-12 浙江大学 基于偏振信息检测野外水体障碍物的方法
JP4157790B2 (ja) * 2003-03-31 2008-10-01 名古屋電機工業株式会社 車両用路面状態検出装置、車両用路面状態検出方法および車両用路面状態検出装置の制御プログラム
CN101610357A (zh) * 2008-06-18 2009-12-23 株式会社理光 摄像装置及路面状态判别方法
CN102901489A (zh) * 2011-07-25 2013-01-30 中兴通讯股份有限公司 路面积水积冰检测方法及装置
CN105518490A (zh) * 2014-12-04 2016-04-20 深圳市大疆创新科技有限公司 一种物体检测方法、装置及遥控移动设备、飞行器

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106802215A (zh) * 2015-11-20 2017-06-06 沈阳新松机器人自动化股份有限公司 一种水管漏水检测装置及检测方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4157790B2 (ja) * 2003-03-31 2008-10-01 名古屋電機工業株式会社 車両用路面状態検出装置、車両用路面状態検出方法および車両用路面状態検出装置の制御プログラム
JP2006058122A (ja) * 2004-08-19 2006-03-02 Nagoya Electric Works Co Ltd 路面状態判別方法およびその装置
CN101033961A (zh) * 2007-02-06 2007-09-12 浙江大学 基于偏振信息检测野外水体障碍物的方法
CN101610357A (zh) * 2008-06-18 2009-12-23 株式会社理光 摄像装置及路面状态判别方法
CN102901489A (zh) * 2011-07-25 2013-01-30 中兴通讯股份有限公司 路面积水积冰检测方法及装置
CN105518490A (zh) * 2014-12-04 2016-04-20 深圳市大疆创新科技有限公司 一种物体检测方法、装置及遥控移动设备、飞行器

Also Published As

Publication number Publication date
CN108701356A (zh) 2018-10-23

Similar Documents

Publication Publication Date Title
US11030470B2 (en) Apparatus and method with liveness verification
CN107209849B (zh) 眼睛跟踪
US10152634B2 (en) Methods and systems for contextually processing imagery
US20190186901A1 (en) Methods and Apparatus for Enhancing Depth Maps with Polarization Cues
US10169637B2 (en) On-screen optical fingerprint capture for user authentication
WO2018119734A1 (zh) 一种显示屏的控制方法及装置
WO2020019760A1 (zh) 活体检测方法、装置及系统、电子设备和存储介质
CN108846837B (zh) 物体表面缺陷检测方法和装置
US9589186B2 (en) Iris recognition apparatus and operating method thereof
US11200436B2 (en) Spoofing detection apparatus, spoofing detection method, and computer-readable recording medium
US11620860B2 (en) Spoofing detection apparatus, spoofing detection method, and computer-readable recording medium
KR102317180B1 (ko) 3차원 깊이정보 및 적외선정보에 기반하여 생체여부의 확인을 행하는 얼굴인식 장치 및 방법
KR102488663B1 (ko) 깊이 프로세서 및 3차원 이미지 기기
KR101919090B1 (ko) 3차원 깊이정보 및 적외선정보에 기반하여 생체여부의 확인을 행하는 얼굴인식 장치 및 방법
US20160283781A1 (en) Display device, display method, and display program
US20030174237A1 (en) Digital stereograph-capturing device
WO2019000313A1 (zh) 一种检测方法、检测设备以及飞行器
JP2019505869A (ja) 複屈折ベースの生体認証のための方法及び装置
US9948825B2 (en) Processing apparatus that can be transitioned into an enabled state
WO2020054076A1 (ja) 情報処理装置および物体検出方法
CN111665932B (zh) 头戴式显示装置以及其眼球追踪装置
CN109886166A (zh) 基于偏振特性的防伪方法和装置
US9232132B1 (en) Light field image processing
JP6972428B2 (ja) 滞在判定装置、滞在判定方法及び滞在判定プログラム
JP2848408B2 (ja) 移動車の環境認識装置

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: 17915790

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17915790

Country of ref document: EP

Kind code of ref document: A1