WO2019137348A1 - 体感摄像头的成像精度的调节方法及调节装置 - Google Patents

体感摄像头的成像精度的调节方法及调节装置 Download PDF

Info

Publication number
WO2019137348A1
WO2019137348A1 PCT/CN2019/070760 CN2019070760W WO2019137348A1 WO 2019137348 A1 WO2019137348 A1 WO 2019137348A1 CN 2019070760 W CN2019070760 W CN 2019070760W WO 2019137348 A1 WO2019137348 A1 WO 2019137348A1
Authority
WO
WIPO (PCT)
Prior art keywords
speckle
infrared
actual
preset
brightness
Prior art date
Application number
PCT/CN2019/070760
Other languages
English (en)
French (fr)
Inventor
周晓军
李骊
王行
盛赞
李朔
Original Assignee
南京华捷艾米软件科技有限公司
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 南京华捷艾米软件科技有限公司 filed Critical 南京华捷艾米软件科技有限公司
Priority to US16/641,653 priority Critical patent/US10986289B2/en
Publication of WO2019137348A1 publication Critical patent/WO2019137348A1/zh

Links

Images

Classifications

    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/10—Image acquisition
    • G06V10/12—Details of acquisition arrangements; Constructional details thereof
    • G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
    • G06V10/145—Illumination specially adapted for pattern recognition, e.g. using gratings
    • 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
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/10—Image acquisition
    • G06V10/12—Details of acquisition arrangements; Constructional details thereof
    • G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
    • G06V10/143—Sensing or illuminating at different wavelengths
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/761—Proximity, similarity or dissimilarity measures
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/20—Cameras or camera modules comprising electronic image sensors; Control thereof for generating image signals from infrared radiation only
    • 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
    • H04N23/67—Focus control based on electronic image sensor signals
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/50—Control of the SSIS exposure
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/70—SSIS architectures; Circuits associated therewith
    • H04N25/76—Addressed sensors, e.g. MOS or CMOS sensors
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00—Details of television systems
    • H04N5/30—Transforming light or analogous information into electric information
    • H04N5/33—Transforming infrared radiation

Definitions

  • the invention relates to the field of somatosensory technology application, in particular to a method for adjusting imaging precision of a somatosensory camera and an adjusting device for imaging precision of a somatosensory camera.
  • the development of somatosensory cameras provides a new way of human-computer interaction.
  • the novel manipulation method controlled by the human body injects new blood into the daily human-computer interaction.
  • some of the sensory camera technologies on the market basically use the infrared transmitting and receiving modules, and the definition of the infrared image determines the quality of the depth map, and finally is reflected in the application of the skeleton recognition, gesture recognition and the like.
  • the accuracy of the infrared image is the first hurdle of the excellent sense of manipulation.
  • infrared CMOS camera modules With the continuous development of infrared CMOS technology in recent years, the manufacturing process of infrared CMOS camera modules has also been continuously changed, which also improves the operability of the sense of body manipulation from another aspect, and brings convenience to the development of somatosensory technology. At the same time, it also has different degrees of influence on the capture of the infrared image. The difference in the design of the somatosensory device itself and the direct combination with the infrared CMOS camera module bring about the problem of excellent discrimination of the infrared image of the CMOS imaging.
  • the present invention aims to at least solve one of the technical problems existing in the prior art, and proposes a method for adjusting the imaging accuracy of a somatosensory camera and an apparatus for adjusting the imaging accuracy of the somatosensory camera.
  • a first aspect of the present invention provides a method for adjusting an imaging accuracy of a somatosensory camera, the method comprising:
  • step S130 Compare the actual definition with the preset definition, and when the actual definition is consistent with the preset definition, proceed to step S140, when the actual definition is inconsistent with the preset definition. , adjusting the imaging focal length of the somatosensory camera, and repeating steps S110 to S130;
  • step S140 comparing the actual speckle regularity with the preset speckle regularity, and when the actual speckle regularity is consistent with the preset speckle regularity, proceeding to step S150, when the actual When the speckle regularity is inconsistent with the preset speckle regularity, the imaging focal length of the somatosensory camera is adjusted, and steps S110 to S140 are repeatedly performed;
  • the method further comprises performing before the step S110:
  • a light source unit is disposed to emit infrared light to the target infrared scene; wherein the light source unit includes at least one light source.
  • the light source comprises a structured light source.
  • the step S120 includes:
  • the step S130 includes:
  • the step S120 includes:
  • the step S140 includes:
  • the distribution of speckles and/or the shape of the speckles in each target detection area with the distribution of the preset speckles and/or the shape of the pre-set speckles, the distribution of speckles in each target detection area and/or Or the shape of the speckle coincides with the distribution of the preset speckle and/or the shape of the preset speckle, and the actual speckle regularity is determined to be consistent with the preset speckle regularity.
  • the step S120 includes:
  • the step S150 includes:
  • an apparatus for adjusting an imaging accuracy of a somatosensory camera comprising an infrared CMOS camera module, the adjusting device comprising an identification module, an infrared speckle clarity module, and infrared speckle regularity Module, infrared speckle center brightness module and imaging precision adjustment module;
  • the infrared CMOS camera module is configured to acquire an infrared speckle pattern in a target infrared scene
  • the identification module is configured to identify an actual sharpness of the infrared speckle pattern, an actual speckle regularity, and an actual brightness of the central region based on the infrared speckle pattern;
  • the infrared speckle clarity module is configured to compare the actual sharpness with a preset sharpness, and send the infrared sharpness regularity module to the infrared speckle regularity module when the actual sharpness is consistent with the preset sharpness a sharpness matching signal, when the actual sharpness is inconsistent with the preset sharpness, sending a sharpness mismatch signal to the imaging precision adjusting module;
  • the infrared speckle regularity module is configured to compare the actual speckle regularity with a preset speckle regularity, and when the actual speckle regularity is consistent with the preset speckle regularity, The infrared speckle center brightness module sends a speckle regularity matching signal, and when the actual speckle regularity is inconsistent with the preset speckle regularity, sending a speckle regularity degree mismatch signal to the imaging precision adjustment module;
  • the infrared speckle center brightness module is configured to compare the actual brightness of the central area with a preset brightness, and complete imaging of the infrared CMOS camera module when the actual brightness of the central area is consistent with the preset brightness. Adjusting the accuracy; when the actual brightness of the central area is inconsistent with the preset brightness, sending a brightness mismatch signal to the imaging precision adjustment module;
  • the imaging precision adjustment module is configured to adjust an imaging focal length of the infrared CMOS camera module according to the definition mismatch signal, the speckle regularity mismatch signal, and the brightness mismatch signal.
  • the somatosensory camera further comprises an infrared emitting module, the infrared emitting module comprising a light source unit to emit infrared light to the target infrared scene; wherein the light source unit comprises at least one light source.
  • the infrared emitting module comprising a light source unit to emit infrared light to the target infrared scene; wherein the light source unit comprises at least one light source.
  • the identification module is configured to identify a grayscale change rate of an edge region of the infrared speckle pattern based on the infrared speckle pattern, and use the grayscale change rate as the actual sharpness;
  • the identification module is configured to divide the infrared speckle pattern into at least one target detection area, and identify a distribution of speckles and/or a shape of a speckle in each of the target detection areas, in the target detection area.
  • the distribution of speckles and/or the shape of the speckle as the actual speckle regularity;
  • the identification module is configured to extract a predetermined range of infrared speckle patterns in the central region
  • An overall luminance average of a predetermined range of infrared speckle patterns in the central region is calculated, and the overall luminance average is used as the actual luminance of the central region.
  • the infrared speckle clarity module is configured to compare a grayscale change rate of the edge region with a preset grayscale change rate, when the grayscale change rate of the edge region is consistent with the preset grayscale change rate. Determining that the actual sharpness is consistent with the preset sharpness;
  • the infrared speckle regularity module is configured to compare the distribution of speckles and/or the shape of the speckles in each target detection area with the distribution of the preset speckles and/or the shape of the preset speckles. Determining the actual speckle regularity and the preset speckle regularity when the distribution of speckles in the detection area and/or the shape of the speckle coincides with the distribution of the preset speckle and/or the shape of the preset speckle Consistent
  • the infrared speckle center brightness module is configured to compare the overall brightness average value with a preset area brightness threshold, and determine the actual brightness of the center area when the overall brightness average value is consistent with the preset area brightness threshold. Consistent with the preset brightness.
  • the method for adjusting the imaging accuracy of the somatosensory camera of the present invention by using the actual sharpness of the infrared speckle pattern in the acquired target infrared scene, the actual speckle regularity and the actual brightness of the central region, and according to the actual definition, the actual dispersion
  • the plaque regularity and the actual brightness of the central region adjust the imaging focal length of the somatosensory camera to obtain an infrared speckle pattern that meets the imaging accuracy requirements, thereby improving the quality of the depth map after the infrared speckle pattern is converted to the depth map for subsequent Skeleton recognition, gesture recognition provides a high quality data source.
  • the imaging precision adjusting device of the somatosensory camera of the present invention passes the actual sharpness of the infrared speckle pattern in the acquired target infrared scene, the actual speckle regularity and the actual brightness of the central region, and according to the actual definition, the actual dispersion
  • the plaque regularity and the actual brightness of the central region adjust the imaging focal length of the somatosensory camera to obtain an infrared speckle pattern that meets the imaging accuracy requirements, thereby improving the quality of the depth map after the infrared speckle pattern is converted to the depth map for subsequent Skeleton recognition, gesture recognition provides a high quality data source.
  • FIG. 1 is a flowchart of a method for adjusting imaging accuracy of a somatosensory camera according to a first embodiment of the present invention
  • FIG. 2 is a schematic structural view of an apparatus for adjusting imaging accuracy of a somatosensory camera according to a second embodiment of the present invention.
  • a first aspect of the present invention relates to a method S100 for adjusting imaging accuracy of a somatosensory camera, the method S100 comprising:
  • step S130 Compare the actual definition with the preset definition, and when the actual definition is consistent with the preset definition, proceed to step S140, when the actual definition is inconsistent with the preset definition. Adjusting the imaging focal length of the somatosensory camera (for example, adjusting the imaging focal length of the infrared CMOS module described below), and repeating steps S110 to S130.
  • step S140 comparing the actual speckle regularity with the preset speckle regularity, and when the actual speckle regularity is consistent with the preset speckle regularity, proceeding to step S150, when the actual When the speckle regularity does not coincide with the preset speckle regularity, the imaging focal length of the somatosensory camera is adjusted, and steps S110 to S140 are repeatedly performed.
  • the method for adjusting the imaging accuracy of the somatosensory camera of the present embodiment by using the actual sharpness of the infrared speckle pattern in the acquired target infrared scene, the actual speckle regularity and the actual brightness of the central region, and according to the actual definition, The actual speckle regularity and the actual brightness of the central region adjust the imaging focal length of the somatosensory camera to obtain an infrared speckle pattern that meets the imaging accuracy requirements, thereby improving the quality of the depth map after the infrared speckle map is converted to the depth map.
  • gesture recognition provides a high quality data source.
  • the method S100 further includes before the step S110:
  • a light source unit is disposed to emit infrared light to the target infrared scene; wherein the light source unit includes at least one light source.
  • the light source comprises a structured light source.
  • the step S120 includes:
  • a grayscale change rate of an edge region of the infrared speckle pattern is identified, and the grayscale change rate is used as the actual sharpness.
  • step S130 includes:
  • the blurred speckle pattern has a very large grayscale at the edge, and the clear infrared speckle pattern is excessively smooth on the edge.
  • This step can directly discriminate the difference in lens focal length. A large situation produces a lens that does not meet the requirements. Therefore, the infrared speckle pattern conforming to the imaging accuracy requirement can be further obtained, thereby improving the quality of the depth map after the infrared speckle pattern is transferred to the depth map, and providing a high-quality data source for subsequent skeleton recognition and gesture recognition.
  • the step S120 includes:
  • the infrared speckle pattern is divided into at least one target detection area.
  • a distribution of speckles and/or a shape of the speckles in each of the target detection regions is identified, with the distribution of speckles and/or the shape of the speckles within the target detection region as the actual speckle regularity.
  • step S140 includes:
  • the distribution of speckles and/or the shape of the speckles in each target detection area with the distribution of the preset speckles and/or the shape of the pre-set speckles, the distribution of speckles in each target detection area and/or Or the shape of the speckle coincides with the distribution of the preset speckle and/or the shape of the preset speckle, and the actual speckle regularity is determined to be consistent with the preset speckle regularity.
  • the infrared speckle regularity check in the infrared CMOS camera module is to further determine the currently taken infrared speckle pattern, if the target The degree of regularity of the speckle in the detection area is large, so the lens focal length under the requirement of sharpness will need to be fine-tuned, because the regularity of the speckle is used to check the detection of the speckle pattern (greater than 1 m) in the near-field area, too The near speckle pattern can not be effectively discriminated because the speckle is too concentrated.
  • the target detection area itself can be freely set, but the speckle regularity scene used for checking should provide rich scene elements as much as possible, and it is not recommended to use an empty scene as the target detection area, in the observation area.
  • the distribution of speckles and the shape of the speckles If the distribution of speckles is uniform and the brightness distribution is reasonable, it can be assumed that the focal length of the infrared lens meets a reasonable set value. In this way, the infrared speckle pattern conforming to the imaging precision requirement can be further obtained, thereby improving the quality of the depth map after the infrared speckle pattern is transferred to the depth map, and providing a high-quality data source for subsequent skeleton recognition and gesture recognition.
  • the step S120 includes:
  • a predetermined range of infrared speckle patterns in the central region is extracted.
  • An overall luminance average of a predetermined range of infrared speckle patterns in the central region is calculated, and the overall luminance average is used as the actual luminance of the central region.
  • the step S150 includes:
  • the infrared central region brightness determination in the infrared CMOS camera module is to detect the speckle pattern of the current image center region, because the central region is scattered.
  • the pattern is in an ideal state, a pattern with a relatively uniform brightness is produced, and when the focal length of the lens is not ideal, for example, a lens focal length that is too far or too close, the center bright spot is very large or the center bright spot is very high.
  • the focal length of the lens is not ideal, for example, a lens focal length that is too far or too close, the center bright spot is very large or the center bright spot is very high.
  • neither of them meets the high quality requirements of infrared speckle.
  • the imaging focal length of the infrared CMOS camera module can be adjusted, and a clearer infrared speckle pattern can be obtained, so that the infrared speckle pattern conforming to the imaging accuracy requirement can be further obtained, thereby passing through the infrared scattering.
  • the bitmap is transferred to the depth map, the quality of the depth map is improved, and a high-quality data source is provided for subsequent skeleton recognition and gesture recognition.
  • the predetermined range may be a fixed size of a central area of the infrared speckle pattern, such as a size of 150 ⁇ 150 pixels, performing luminance statistics of the central area, and setting a reasonable threshold when the calculated area value satisfies one
  • the required value is not returned to adjust the focal length of the infrared CMOS camera module, otherwise the focus is continuously adjusted until the set predetermined area brightness threshold is met.
  • the infrared speckle pattern of the same position region in the same scene and the infrared speckle pattern of the same environment can be taken, and the average brightness value of the area is calculated as the brightness of the infrared speckle pattern.
  • the value is generally decentered. Because the focal length deviation is large in the central region, the brightness of the central region will become very bright, and the brightness of the central region of the ideal infrared speckle pattern will not increase sharply. There will be no very large bright spots in the central area, and a reasonable threshold is generally recommended to be about 10% of the scene threshold.
  • the method for adjusting the imaging accuracy of the somatosensory camera of the present invention is simpler and more effective.
  • the invention adopts the definition based on the sharpness discrimination, the speckle rule degree discrimination, and the central region brightness discrimination statistics, and can be very intuitively used to indicate whether the infrared CMOS camera module has reached a very ideal state, and whether the focal length of the lens satisfies the needs of the scene.
  • the settings provide a more accurate source of data for subsequent depth calculations.
  • the invention proposes a resolution judging step, and calculating an edge gray gradation change rate at which each row of consecutive descending interval pixels in an image represents a gray gradation change rate of the line, and finally calculates a gradation change rate of the entire image.
  • the invention proposes speckle rule degree discrimination, selects a certain area of the entire infrared speckle central area to perform speckle rule statistics, and therefore, the imaging precision adjustment method of the invention can be improved after the infrared speckle map is rotated to the depth map.
  • the quality of the depth map provides a high quality data source for subsequent skeleton recognition and gesture recognition.
  • an adjustment device 200 for imaging accuracy of a somatosensory camera is provided.
  • the somatosensory camera 100 includes an infrared CMOS camera module 110, and the adjustment device includes an identification module 210 and an infrared ray.
  • the infrared CMOS camera module 110 is configured to acquire an infrared speckle pattern in a target infrared scene
  • the identification module 210 is configured to identify an actual sharpness of the infrared speckle pattern, an actual speckle regularity, and an actual brightness of the central region based on the infrared speckle pattern;
  • the infrared speckle clarity module 220 is configured to compare the actual sharpness with a preset sharpness, and when the actual sharpness is consistent with the preset sharpness, to the infrared speckle regularity module 230 transmitting a sharpness matching signal, and sending a sharpness mismatch signal to the imaging precision adjusting module 250 when the actual sharpness is inconsistent with the preset sharpness;
  • the infrared speckle regularity module 230 is configured to compare the actual speckle regularity with a preset speckle regularity, and when the actual speckle regularity is consistent with the preset speckle regularity, Sending a speckle regularity matching signal to the infrared speckle center luminance module 240, and when the actual speckle regularity is inconsistent with the preset speckle regularity, sending the speckle regularity to the imaging precision adjustment module 250 Matching signal
  • the infrared speckle center brightness module 240 is configured to compare the actual brightness of the central area with a preset brightness, and complete the infrared CMOS camera module 110 when the actual brightness of the center area is consistent with the preset brightness. Adjusting the imaging accuracy; when the actual brightness of the central area is inconsistent with the preset brightness, sending a brightness mismatch signal to the imaging precision adjustment module 250;
  • the imaging precision adjustment module 250 is configured to adjust an imaging focal length of the infrared CMOS camera module 110 according to the definition mismatch signal, the speckle regularity mismatch signal, and the brightness mismatch signal.
  • the imaging precision adjusting apparatus 100 of the somatosensory camera of the embodiment passes the actual sharpness of the infrared speckle pattern in the acquired target infrared scene, the actual speckle regularity and the actual brightness of the central area, and according to the actual definition, The actual speckle regularity and the actual brightness of the central region adjust the imaging focal length of the somatosensory camera to obtain an infrared speckle pattern that meets the imaging accuracy requirements, thereby improving the quality of the depth map after the infrared speckle map is converted to the depth map.
  • Subsequent skeleton recognition, gesture recognition provides a high quality data source.
  • the somatosensory camera 100 further includes an infrared emitting module 120, and the infrared emitting module 120 includes a light source unit (not shown) to emit infrared light to the target infrared scene;
  • the light source unit comprises at least one light source (not shown).
  • the identification module 210 is configured to identify a grayscale change rate of an edge region of the infrared speckle pattern based on the infrared speckle pattern, and use the grayscale change rate as the actual sharpness; and
  • the identification module 210 is configured to divide the infrared speckle pattern into at least one target detection area, and identify a distribution of speckles and/or a shape of a speckle in each of the target detection areas, in the target detection area.
  • the distribution of speckles and/or the shape of the speckle as the actual speckle regularity; and,
  • the identification module 210 is configured to extract a predetermined range of infrared speckle patterns in the central region;
  • An overall luminance average of a predetermined range of infrared speckle patterns in the central region is calculated, and the overall luminance average is used as the actual luminance of the central region.
  • the infrared speckle clarity module 220 is configured to compare the grayscale change rate of the edge region with a preset grayscale change rate, when the grayscale change rate of the edge region is consistent with the preset grayscale change rate. Determining that the actual sharpness is consistent with the preset sharpness;
  • the infrared speckle regularity module 230 is configured to compare the distribution of speckles and/or the shape of the speckles in each target detection area with the distribution of the preset speckles and/or the shape of the preset speckles. Determining the actual speckle regularity and the preset speckle rule when the distribution of speckles in the target detection area and/or the shape of the speckle coincides with the distribution of the preset speckle and/or the shape of the preset speckle Consistent
  • the infrared speckle center brightness module 240 is configured to compare the overall brightness average value with a preset area brightness threshold, and determine that the central area is actually actual when the overall brightness average value is consistent with a preset area brightness threshold. The brightness is consistent with the preset brightness.
  • the infrared emitting module 120 emits infrared light through several different processing modules before being emitted into the space.
  • an invisible infrared light source passes through a collimating lens (WLO).
  • the emitted infrared source passes through the diffraction source (DOE) and is projected into the spatial field of view of a rectangular cone.
  • DOE diffraction source
  • the DOE-diffracted source produces random infrared speckles. These regular speckles meet differently in space.
  • the CMOS infrared camera module 110 captures the infrared speckles in these spaces and presents them on the infrared speckle pattern.
  • different rules of speckle on different objects, and the rule program of speckle is another proof of the reasonable focal length of an infrared lens.
  • the adjustment device 100 for imaging accuracy of the somatosensory camera of the present invention has a simpler structure and more effective effect.
  • the invention adopts the definition based on the sharpness discrimination, the speckle rule degree discrimination, and the central region brightness discrimination statistics, and can be very intuitively used to indicate whether the infrared CMOS camera module has reached a very ideal state, and whether the focal length of the lens satisfies the needs of the scene.
  • the settings provide a more accurate source of data for subsequent depth calculations.
  • the specific resolution method, the speckle rule degree discriminating method, and the central region brightness discriminating method reference may be made to the related descriptions herein, and no further description is provided herein. Therefore, with the imaging precision adjusting device of the present invention, after the infrared speckle map is rotated to the depth map, the quality of the depth map can be improved, and a high quality data source is provided for subsequent skeleton recognition and gesture recognition.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • General Engineering & Computer Science (AREA)
  • Studio Devices (AREA)
  • Photometry And Measurement Of Optical Pulse Characteristics (AREA)

Abstract

本发明公开了一种体感摄像头的成像精度的调节方法及装置，。所述方法包括获取目标红外场景中的红外散斑图，识别所述红外散斑图的实际清晰度、实际散斑规则度和中心区域实际亮度；分别将所述实际清晰度与预设清晰度进行比较、将所述实际散斑规则度与预设散斑规则度进行比较，以及将所述中心区域实际亮度与预设亮度进行比较，并且，根据比较结果调节成像焦距，完成成像精度的调节。因此，利用本发明的体感摄像头的成像精度的调节方法，以此可以获得符合成像精度要求的红外散斑图，从而经过红外散斑图转深度图后，提高深度图的质量，为后续的骨架识别，手势识别提供高质量的数据源。

Description

体感摄像头的成像精度的调节方法及调节装置 技术领域
本发明涉及体感技术应用领域，特别涉及一种体感摄像头的成像精度的调节方法和一种体感摄像头的成像精度的调节装置。
背景技术
目前，体感摄像头的发展提供了一种人机交互的全新方式，人体操控的新奇操控方法给日常的人机交互方式注入了新的血液。目前市场上部分的体感摄像头技术中基本都用到了红外的发射与接收的模块，并且红外图的清晰度决定了深度图质量的高低，并最终体现在了骨架的识别，手势的识别等应用领域上，可以说，红外图的精度高低，是优秀的体感操控方式的第一道坎。
而随着近几年红外CMOS技术的不断发展，红外CMOS摄像模块的制作工艺亦不断的发展改变，这也给体感的操控从另一个方面提高了可操作性，给体感技术发展带来便利的同时，也对红外图的捕捉上带来了不同程度的影响，体感设备本身的设计的差异性以及与红外CMOS摄像模块直接的结合带来了CMOS成像的红外图精度优异判别的问题。
因此，如何设计出一种根据红外散斑图调节体感摄像头的成像精度成为本领域亟需解决的技术问题。
发明内容
本发明旨在至少解决现有技术中存在的技术问题之一，提出了一种体感摄像头的成像精度的调节方法和一种体感摄像头的成像精度的调节装置。
为了实现上述目的，本发明的第一方面，提供了一种体感摄像头的成像精度的调节方法，所述方法包括：
S110、获取目标红外场景中的红外散斑图；
S120、基于所述红外散斑图，识别所述红外散斑图的实际清晰度、实际散斑规则度和中心区域实际亮度；
S130、将所述实际清晰度与预设清晰度进行比较，并且，当所述实际清晰度与预设清晰度一致时，转入步骤S140，当所述实际清晰度与预设清晰度不一致时，调节体感摄像头的成像焦距，并重复执行步骤S110至步骤S130；
S140、将所述实际散斑规则度与预设散斑规则度进行比较，并且，当所述实际散斑规则度与预设散斑规则度相一致时，转入步骤S150，当所述实际散斑规则度与预设散斑规则度不一致时，调节体感摄像头的成像焦距，并重复执行步骤S110至步骤S140；
S150、将所述中心区域实际亮度与预设亮度进行比较，并且，当所述中心区域实际亮度与预设亮度相一致时，完成所述体感摄像头的成像精度的调节；当所述中心区域实际亮度与预设亮度不一致时，调节体感摄像头的成像焦距，并重复执行步骤S110至步骤S150。
优选地，所述方法还包括在所述步骤S110之前进行的：
S101、设置光源单元，以向所述目标红外场景发射红外光；其中，所述光源单元包括至少一个光源。
优选地，所述光源包括结构光光源。
优选地，所述步骤S120包括：
基于所述红外散斑图，识别所述红外散斑图的边缘区域的灰度变化率，以所述灰度变化率作为所述实际清晰度；
所述步骤S130包括：
将边缘区域的灰度变化率与预设灰度变化率进行比较，当所述边缘区域的灰度变化率与预设灰度变化率一致时，判定所述实际清晰度与预设清晰度相一致。
优选地，所述步骤S120包括：
将所述红外散斑图划分为至少一个目标检测区域；
识别各所述目标检测区域内的散斑的分布和/或散斑的形状，以所述目标检测区域内的散斑的分布和/或散斑的形状作为所述实际散 斑规则度；
所述步骤S140包括：
将各目标检测区域内的散斑的分布和/或散斑的形状与预设散斑的分布和/或预设散斑的形状进行比较，当各目标检测区域内的散斑的分布和/或散斑的形状与预设散斑的分布和/或预设散斑的形状相一致时，判定所述实际散斑规则度与预设散斑规则度相一致。
优选地，所述步骤S120包括：
提取中心区域内预定范围的红外散斑图；
计算所述中心区域内预定范围的红外散斑图的整体亮度平均值，以所述整体亮度平均值作为所述中心区域实际亮度；
所述步骤S150包括：
将所述整体亮度平均值与预设区域亮度阈值进行比较，当所述整体亮度平均值与预设区域亮度阈值相一致时，判定所述中心区域实际亮度与预设亮度相一致。
本发明的第二方面，提供了一种体感摄像头的成像精度的调节装置，所述体感摄像头包括红外CMOS摄像模块，所述调节装置包括识别模块、红外散斑清晰度模块、红外散斑规则度模块、红外散斑中心亮度模块和成像精度调节模块；
所述红外CMOS摄像模块用于获取目标红外场景中的红外散斑图；
所述识别模块用于基于所述红外散斑图，识别所述红外散斑图的实际清晰度、实际散斑规则度和中心区域实际亮度；
所述红外散斑清晰度模块用于将所述实际清晰度与预设清晰度进行比较，并且，当所述实际清晰度与预设清晰度一致时，向所述红外散斑规则度模块发送清晰度匹配信号，当所述实际清晰度与预设清晰度不一致时，向所述成像精度调节模块发送清晰度不匹配信号；
所述红外散斑规则度模块用于将所述实际散斑规则度与预设散斑规则度进行比较，并且，当所述实际散斑规则度与预设散斑规则度相一致时，向所述红外散斑中心亮度模块发送散斑规则度匹配信号，当所述实际散斑规则度与预设散斑规则度不一致时，向所述成像精度 调节模块发送散斑规则度不匹配信号；
所述红外散斑中心亮度模块用于将所述中心区域实际亮度与预设亮度进行比较，并且，当所述中心区域实际亮度与预设亮度相一致时，完成所述红外CMOS摄像模块的成像精度的调节；当所述中心区域实际亮度与预设亮度不一致时，向所述成像精度调节模块发送亮度不匹配信号；
所述成像精度调节模块用于根据所述清晰度不匹配信号、散斑规则度不匹配信号和所述亮度不匹配信号，调节所述红外CMOS摄像模块的成像焦距。
优选地，所述体感摄像头还包括红外发射模块，所述红外发射模块包括光源单元，以向所述目标红外场景发射红外光；其中，所述光源单元包括至少一个光源。
优选地，所述识别模块用于基于所述红外散斑图，识别所述红外散斑图的边缘区域的灰度变化率，以所述灰度变化率作为所述实际清晰度；以及，
所述识别模块用于将所述红外散斑图划分为至少一个目标检测区域，识别各所述目标检测区域内的散斑的分布和/或散斑的形状，以所述目标检测区域内的散斑的分布和/或散斑的形状作为所述实际散斑规则度；以及，
所述识别模块用于提取中心区域内预定范围的红外散斑图；
计算所述中心区域内预定范围的红外散斑图的整体亮度平均值，以所述整体亮度平均值作为所述中心区域实际亮度。
优选地，所述红外散斑清晰度模块用于将边缘区域的灰度变化率与预设灰度变化率进行比较，当所述边缘区域的灰度变化率与预设灰度变化率一致时，判定所述实际清晰度与预设清晰度相一致；
所述红外散斑规则度模块用于将各目标检测区域内的散斑的分布和/或散斑的形状与预设散斑的分布和/或预设散斑的形状进行比较，当各目标检测区域内的散斑的分布和/或散斑的形状与预设散斑的分布和/或预设散斑的形状相一致时，判定所述实际散斑规则度与预设散斑规则度相一致；
所述红外散斑中心亮度模块用于将所述整体亮度平均值与预设区域亮度阈值进行比较，当所述整体亮度平均值与预设区域亮度阈值相一致时，判定所述中心区域实际亮度与预设亮度相一致。
本发明的体感摄像头的成像精度的调节方法，通过对获取到的目标红外场景中的红外散斑图的实际清晰度、实际散斑规则度和中心区域实际亮度，并根据实际清晰度、实际散斑规则度和中心区域实际亮度对体感摄像头的成像焦距进行调节，以此获得符合成像精度要求的红外散斑图，从而经过红外散斑图转深度图后，提高深度图的质量，为后续的骨架识别，手势识别提供高质量的数据源。
本发明的体感摄像头的成像精度的调节装置，通过对获取到的目标红外场景中的红外散斑图的实际清晰度、实际散斑规则度和中心区域实际亮度，并根据实际清晰度、实际散斑规则度和中心区域实际亮度对体感摄像头的成像焦距进行调节，以此获得符合成像精度要求的红外散斑图，从而经过红外散斑图转深度图后，提高深度图的质量，为后续的骨架识别，手势识别提供高质量的数据源。
附图说明
附图是用来提供对本发明的进一步理解，并且构成说明书的一部分，与下面的具体实施方式一起用于解释本发明，但并不构成对本发明的限制。在附图中：
图1为本发明第一实施例中体感摄像头的成像精度的调节方法的流程图；
图2为本发明第二实施例中体感摄像头的成像精度的调节装置的结构示意图。
附图标记说明
200：体感摄像头的成像精度的调节装置；
210：识别模块；
220：红外散斑清晰度模块；
230：红外散斑规则度模块；
240：红外散斑中心亮度模块；
250：成像精度调节模块；
100：体感摄像头；
110：红外CMOS摄像模块；
120：红外发射模块。
具体实施方式
以下结合附图对本发明的具体实施方式进行详细说明。应当理解的是，此处所描述的具体实施方式仅用于说明和解释本发明，并不用于限制本发明。
参考图1，本发明的第一方面，涉及一种体感摄像头的成像精度的调节方法S100，所述方法S100包括：
S110、获取目标红外场景中的红外散斑图。
S120、基于所述红外散斑图，识别所述红外散斑图的实际清晰度、实际散斑规则度和中心区域实际亮度。
S130、将所述实际清晰度与预设清晰度进行比较，并且，当所述实际清晰度与预设清晰度一致时，转入步骤S140，当所述实际清晰度与预设清晰度不一致时，调节体感摄像头的成像焦距(例如，调节下述红外CMOS模组的成像焦距)，并重复执行步骤S110至步骤S130。
S140、将所述实际散斑规则度与预设散斑规则度进行比较，并且，当所述实际散斑规则度与预设散斑规则度相一致时，转入步骤S150，当所述实际散斑规则度与预设散斑规则度不一致时，调节体感摄像头的成像焦距，并重复执行步骤S110至步骤S140。
S150、将所述中心区域实际亮度与预设亮度进行比较，并且，当所述中心区域实际亮度与预设亮度相一致时，完成所述体感摄像头的成像精度的调节；当所述中心区域实际亮度与预设亮度不一致时，调节体感摄像头的成像焦距，并重复执行步骤S110至步骤S150。
本实施例的体感摄像头的成像精度的调节方法S100，通过对获 取到的目标红外场景中的红外散斑图的实际清晰度、实际散斑规则度和中心区域实际亮度，并根据实际清晰度、实际散斑规则度和中心区域实际亮度对体感摄像头的成像焦距进行调节，以此获得符合成像精度要求的红外散斑图，从而经过红外散斑图转深度图后，提高深度图的质量，为后续的骨架识别，手势识别提供高质量的数据源。
优选地，所述方法S100还包括在所述步骤S110之前进行的：
S101、设置光源单元，以向所述目标红外场景发射红外光；其中，所述光源单元包括至少一个光源。为了提高体感摄像头的红外散斑图的成像精度调节，优选地，所述光源包括结构光光源。
优选地，所述步骤S120包括：
基于所述红外散斑图，识别所述红外散斑图的边缘区域的灰度变化率，以所述灰度变化率作为所述实际清晰度。
相应地，所述步骤S130包括：
将边缘区域的灰度变化率与预设灰度变化率进行比较，当所述边缘区域的灰度变化率与预设灰度变化率一致时，判定所述实际清晰度与预设清晰度相一致。
以调节体感摄像头中的红外CMOS摄像模块的成像焦距为例进行说明：
对当前捕获的红外散斑图进行对比，模糊的散斑图其边缘的灰度过度非常大，而清晰的红外散斑图在边缘上过度比较圆滑，这一步可以直接判别出因镜头焦距差别太大情况产生的不符合要求镜头。因此，可以进一步获得符合成像精度要求的红外散斑图，从而经过红外散斑图转深度图后，提高深度图的质量，为后续的骨架识别，手势识别提供高质量的数据源。
优选地，所述步骤S120包括：
将所述红外散斑图划分为至少一个目标检测区域。
识别各所述目标检测区域内的散斑的分布和/或散斑的形状，以所述目标检测区域内的散斑的分布和/或散斑的形状作为所述实际散斑规则度。
相应地，所述步骤S140包括：
将各目标检测区域内的散斑的分布和/或散斑的形状与预设散斑的分布和/或预设散斑的形状进行比较，当各目标检测区域内的散斑的分布和/或散斑的形状与预设散斑的分布和/或预设散斑的形状相一致时，判定所述实际散斑规则度与预设散斑规则度相一致。
具体地，本实施例的体感摄像头的成像精度的调节方法S100中，红外CMOS摄像模块中的红外散斑规则度检查，则是对当前拍摄的红外散斑图进行更近一步的判别，如果目标检测区域内的散斑的规则程度较大，那么在满足清晰度要求下的镜头焦距将需要微调，因为散斑的规则度用来检查近景区域的散斑图形(大于1m)左右的检测，太近的散斑图会因为散斑太过集中，而无法有效判别。
需要说明的是，上述目标检测区域的本身可以自由设定，但用来检查的散斑规则度场景中尽量应提供丰富的场景元素，不建议采用空旷的场景作为目标检测区域，观察区域中的散斑的分布以及散斑的形状，若散斑的分布均匀，且亮度分布较为合理，那么可以认定红外的镜头焦距符合合理的设定值。这样，可以进一步获得符合成像精度要求的红外散斑图，从而经过红外散斑图转深度图后，提高深度图的质量，为后续的骨架识别，手势识别提供高质量的数据源。
优选地，所述步骤S120包括：
提取中心区域内预定范围的红外散斑图。
计算所述中心区域内预定范围的红外散斑图的整体亮度平均值，以所述整体亮度平均值作为所述中心区域实际亮度。
所述步骤S150包括：
将所述整体亮度平均值与预设区域亮度阈值进行比较，当所述整体亮度平均值与预设区域亮度阈值相一致时，判定所述中心区域实际亮度与预设亮度相一致。
具体地，在本实施例的体感摄像头的成像精度的调节方法S100中，红外CMOS摄像模块中的红外中心区域亮度判别，则是对当前图像中心区域的散斑图进行检测，因中心区域的散斑图在较理想的状态下时，产生的比较均匀亮度的图案，而镜头焦距不理想的情况，比如，太远或者太近的镜头焦距，会产生中心亮斑非常大或者中心亮斑 非常的模块情况，这两者皆不满足红外散斑的高质量要求。因此，根据红外散斑图的中心区域的亮度调节红外CMOS摄像模块的成像焦距，可以获得更清晰的红外散斑图，这样，可以进一步获得符合成像精度要求的红外散斑图，从而经过红外散斑图转深度图后，提高深度图的质量，为后续的骨架识别，手势识别提供高质量的数据源。
更具体地，上述预定范围，例如，可以是红外散斑图的中心区域的固定大小，比如150x150个像素点大小，进行中心区域的亮度统计，同时设立一个合理的阈值，当计算区域值满足一个要求的值便不在返回调节红外CMOS摄像模块的焦距，否则不断的调整焦距，直至满足设定的预定区域亮度阈值。
至于预定区域亮度阈值的设定的方法，一般可取同场景，同环境的红外散斑图中的同位置区域的红外散斑图，计算该区域的平均亮度值作为该张红外散斑图的亮度值，一般去中心区域，因中心区域在焦距偏差较大是，中心区域亮度会变得非常的亮，而较为理想的红外散斑图中心区域的亮度较周围亮度衰减梯度不会突增，更不会在中心区域的出现非常大的亮斑，一般建议合理的阈值为场景阈值的10％左右。
因此，本发明的体感摄像头的成像精度的调节方法S100，成像精度的调节方法更加简单，且更有效果。本发明采用基于清晰度判别，散斑规则度判别，以及中心区域亮度判别统计，能非常直观的用来表示红外CMOS摄像模块是否达到了一个非常理想的状态，其镜头的焦距是否满足场景的需要的设定，给后续的深度计算提供了更加精确的数据源。本发明提出清晰度判别步骤，计算图像中每一行连续下降间隔像素点最多的边缘灰度变化率代表了这一行的灰度变化率，并最终计算出整张图的灰度变化率。本发明提出散斑规则度判别，选取整张红外散斑中心区域的一定面积进行散斑规则统计，因此，利用本发明的成像精度调节方法，在经过红外散斑图转深度图后，可以提高深度图的质量，为后续的骨架识别，手势识别提供高质量的数据源。
本发明的第二方面，如图2所示，提供了一种体感摄像头的成像精度的调节装置200，所述体感摄像头100包括红外CMOS摄像模 块110，所述调节装置包括识别模块210、红外散斑清晰度模块220、红外散斑规则度模块230、红外散斑中心亮度模块240和成像精度调节模块250。
所述红外CMOS摄像模块110用于获取目标红外场景中的红外散斑图；
所述识别模块210用于基于所述红外散斑图，识别所述红外散斑图的实际清晰度、实际散斑规则度和中心区域实际亮度；
所述红外散斑清晰度模块220用于将所述实际清晰度与预设清晰度进行比较，并且，当所述实际清晰度与预设清晰度一致时，向所述红外散斑规则度模块230发送清晰度匹配信号，当所述实际清晰度与预设清晰度不一致时，向所述成像精度调节模块250发送清晰度不匹配信号；
所述红外散斑规则度模块230用于将所述实际散斑规则度与预设散斑规则度进行比较，并且，当所述实际散斑规则度与预设散斑规则度相一致时，向所述红外散斑中心亮度模块240发送散斑规则度匹配信号，当所述实际散斑规则度与预设散斑规则度不一致时，向所述成像精度调节模块250发送散斑规则度不匹配信号；
所述红外散斑中心亮度模块240用于将所述中心区域实际亮度与预设亮度进行比较，并且，当所述中心区域实际亮度与预设亮度相一致时，完成所述红外CMOS摄像模块110的成像精度的调节；当所述中心区域实际亮度与预设亮度不一致时，向所述成像精度调节模块250发送亮度不匹配信号；
所述成像精度调节模块250用于根据所述清晰度不匹配信号、散斑规则度不匹配信号和所述亮度不匹配信号，调节所述红外CMOS摄像模块110的成像焦距。
本实施例的体感摄像头的成像精度的调节装置100，通过对获取到的目标红外场景中的红外散斑图的实际清晰度、实际散斑规则度和中心区域实际亮度，并根据实际清晰度、实际散斑规则度和中心区域实际亮度对体感摄像头的成像焦距进行调节，以此获得符合成像精度要求的红外散斑图，从而经过红外散斑图转深度图后，提高深度图的 质量，为后续的骨架识别，手势识别提供高质量的数据源。
优选地，如图2所示，所述体感摄像头100还包括红外发射模块120，所述红外发射模块120包括光源单元(图中并未示出)，以向所述目标红外场景发射红外光；其中，所述光源单元包括至少一个光源(图中并未示出)。
优选地，所述识别模块210用于基于所述红外散斑图，识别所述红外散斑图的边缘区域的灰度变化率，以所述灰度变化率作为所述实际清晰度；以及，
所述识别模块210用于将所述红外散斑图划分为至少一个目标检测区域，识别各所述目标检测区域内的散斑的分布和/或散斑的形状，以所述目标检测区域内的散斑的分布和/或散斑的形状作为所述实际散斑规则度；以及，
所述识别模块210用于提取中心区域内预定范围的红外散斑图；
计算所述中心区域内预定范围的红外散斑图的整体亮度平均值，以所述整体亮度平均值作为所述中心区域实际亮度。
优选地，所述红外散斑清晰度模块220用于将边缘区域的灰度变化率与预设灰度变化率进行比较，当所述边缘区域的灰度变化率与预设灰度变化率一致时，判定所述实际清晰度与预设清晰度相一致；
所述红外散斑规则度模块230用于将各目标检测区域内的散斑的分布和/或散斑的形状与预设散斑的分布和/或预设散斑的形状进行比较，当各目标检测区域内的散斑的分布和/或散斑的形状与预设散斑的分布和/或预设散斑的形状相一致时，判定所述实际散斑规则度与预设散斑规则度相一致；
所述红外散斑中心亮度模块240用于将所述整体亮度平均值与预设区域亮度阈值进行比较，当所述整体亮度平均值与预设区域亮度阈值相一致时，判定所述中心区域实际亮度与预设亮度相一致。
需要说明的是，上述红外发射模块120，在发射出来的红外光经过了几个不同的处理模块才发射到空间中去，如，不可见的红外光源，经过了准直镜头(WLO)，其发出的红外光源经过了衍射源件(DOE)，投射到了一个矩形椎体的空间视野中，而经过DOE衍射的光源，产 生了随机的红外散斑，这些规则的散斑在空间中遇见不同的人与物时，会产生出不同的形状，而不同形状的散斑，其中的CMOS红外摄像模块110在拍摄到这些空间中的红外散斑，呈现在红外散斑图上时，便直观的体现了除了空间中不同地点，不同物体上不同规则的散斑，而散斑的规则程序则是衡量一个红外镜头的焦距合理的另一个佐证。
因此，本发明的体感摄像头的成像精度的调节装置100，成像精度的调节装置结构更加简单，且更有效果。本发明采用基于清晰度判别，散斑规则度判别，以及中心区域亮度判别统计，能非常直观的用来表示红外CMOS摄像模块是否达到了一个非常理想的状态，其镜头的焦距是否满足场景的需要的设定，给后续的深度计算提供了更加精确的数据源。至于具体地清晰度判别方式、散斑规则度判别方式以及中心区域亮度判别方式，可以参考前文相关记载，在此不作赘述。因此，利用本发明的成像精度调节装置，在经过红外散斑图转深度图后，可以提高深度图的质量，为后续的骨架识别，手势识别提供高质量的数据源。
可以理解的是，以上实施方式仅仅是为了说明本发明的原理而采用的示例性实施方式，然而本发明并不局限于此。对于本领域内的普通技术人员而言，在不脱离本发明的精神和实质的情况下，可以做出各种变型和改进，这些变型和改进也视为本发明的保护范围。

Claims (10)

  1. 一种体感摄像头的成像精度的调节方法，其特征在于，所述方法包括：
    S110、获取目标红外场景中的红外散斑图；
    S120、基于所述红外散斑图，识别所述红外散斑图的实际清晰度、实际散斑规则度和中心区域实际亮度；
    S130、将所述实际清晰度与预设清晰度进行比较，并且，当所述实际清晰度与预设清晰度一致时，转入步骤S140，当所述实际清晰度与预设清晰度不一致时，调节体感摄像头的成像焦距，并重复执行步骤S110至步骤S130；
    S140、将所述实际散斑规则度与预设散斑规则度进行比较，并且，当所述实际散斑规则度与预设散斑规则度相一致时，转入步骤S150，当所述实际散斑规则度与预设散斑规则度不一致时，调节体感摄像头的成像焦距，并重复执行步骤S110至步骤S140；
    S150、将所述中心区域实际亮度与预设亮度进行比较，并且，当所述中心区域实际亮度与预设亮度相一致时，完成所述体感摄像头的成像精度的调节；当所述中心区域实际亮度与预设亮度不一致时，调节体感摄像头的成像焦距，并重复执行步骤S110至步骤S150。
  2. 根据权利要求1所述的体感摄像头的成像精度的调节方法，其特征在于，所述方法还包括在所述步骤S110之前进行的：
    S101、设置光源单元，以向所述目标红外场景发射红外光；其中，所述光源单元包括至少一个光源。
  3. 根据权利要求2所述的体感摄像头的成像精度的调节方法，其特征在于，所述光源包括结构光光源。
  4. 根据权利要求1至3中任意一项所述的体感摄像头的成像精度的调节方法，其特征在于，
    所述步骤S120包括：
    基于所述红外散斑图，识别所述红外散斑图的边缘区域的灰度变化率，以所述灰度变化率作为所述实际清晰度；
    所述步骤S130包括：
    将边缘区域的灰度变化率与预设灰度变化率进行比较，当所述边缘区域的灰度变化率与预设灰度变化率一致时，判定所述实际清晰度与预设清晰度相一致。
  5. 根据权利要求1至3中任意一项所述的体感摄像头的成像精度的调节方法，其特征在于，
    所述步骤S120包括：
    将所述红外散斑图划分为至少一个目标检测区域；
    识别各所述目标检测区域内的散斑的分布和/或散斑的形状，以所述目标检测区域内的散斑的分布和/或散斑的形状作为所述实际散斑规则度；
    所述步骤S140包括：
    将各目标检测区域内的散斑的分布和/或散斑的形状与预设散斑的分布和/或预设散斑的形状进行比较，当各目标检测区域内的散斑的分布和/或散斑的形状与预设散斑的分布和/或预设散斑的形状相一致时，判定所述实际散斑规则度与预设散斑规则度相一致。
  6. 根据权利要求1至3中任意一项所述的体感摄像头的成像精度的调节方法，其特征在于，
    所述步骤S120包括：
    提取中心区域内预定范围的红外散斑图；
    计算所述中心区域内预定范围的红外散斑图的整体亮度平均值，以所述整体亮度平均值作为所述中心区域实际亮度；
    所述步骤S150包括：
    将所述整体亮度平均值与预设区域亮度阈值进行比较，当所述整体亮度平均值与预设区域亮度阈值相一致时，判定所述中心区域实 际亮度与预设亮度相一致。
  7. 一种体感摄像头的成像精度的调节装置，其特征在于，所述体感摄像头包括红外CMOS摄像模块，所述调节装置包括识别模块、红外散斑清晰度模块、红外散斑规则度模块、红外散斑中心亮度模块和成像精度调节模块；
    所述红外CMOS摄像模块用于获取目标红外场景中的红外散斑图；
    所述识别模块用于基于所述红外散斑图，识别所述红外散斑图的实际清晰度、实际散斑规则度和中心区域实际亮度；
    所述红外散斑清晰度模块用于将所述实际清晰度与预设清晰度进行比较，并且，当所述实际清晰度与预设清晰度一致时，向所述红外散斑规则度模块发送清晰度匹配信号，当所述实际清晰度与预设清晰度不一致时，向所述成像精度调节模块发送清晰度不匹配信号；
    所述红外散斑规则度模块用于将所述实际散斑规则度与预设散斑规则度进行比较，并且，当所述实际散斑规则度与预设散斑规则度相一致时，向所述红外散斑中心亮度模块发送散斑规则度匹配信号，当所述实际散斑规则度与预设散斑规则度不一致时，向所述成像精度调节模块发送散斑规则度不匹配信号；
    所述红外散斑中心亮度模块用于将所述中心区域实际亮度与预设亮度进行比较，并且，当所述中心区域实际亮度与预设亮度相一致时，完成所述红外CMOS摄像模块的成像精度的调节；当所述中心区域实际亮度与预设亮度不一致时，向所述成像精度调节模块发送亮度不匹配信号；
    所述成像精度调节模块用于根据所述清晰度不匹配信号、散斑规则度不匹配信号和所述亮度不匹配信号，调节所述红外CMOS摄像模块的成像焦距。
  8. 根据权利要求7所述的体感摄像头的成像精度的调节装置，其特征在于，所述体感摄像头还包括红外发射模块，所述红外发射模 块包括光源单元，以向所述目标红外场景发射红外光；其中，所述光源单元包括至少一个光源。
  9. 根据权利要求7或8所述的体感摄像头的成像精度的调节装置，其特征在于，
    所述识别模块用于基于所述红外散斑图，识别所述红外散斑图的边缘区域的灰度变化率，以所述灰度变化率作为所述实际清晰度；以及，
    所述识别模块用于将所述红外散斑图划分为至少一个目标检测区域，识别各所述目标检测区域内的散斑的分布和/或散斑的形状，以所述目标检测区域内的散斑的分布和/或散斑的形状作为所述实际散斑规则度；以及，
    所述识别模块用于提取中心区域内预定范围的红外散斑图；
    计算所述中心区域内预定范围的红外散斑图的整体亮度平均值，以所述整体亮度平均值作为所述中心区域实际亮度。
  10. 根据权利要求9所述的体感摄像头的成像精度的调节装置，其特征在于，
    所述红外散斑清晰度模块用于将边缘区域的灰度变化率与预设灰度变化率进行比较，当所述边缘区域的灰度变化率与预设灰度变化率一致时，判定所述实际清晰度与预设清晰度相一致；
    所述红外散斑规则度模块用于将各目标检测区域内的散斑的分布和/或散斑的形状与预设散斑的分布和/或预设散斑的形状进行比较，当各目标检测区域内的散斑的分布和/或散斑的形状与预设散斑的分布和/或预设散斑的形状相一致时，判定所述实际散斑规则度与预设散斑规则度相一致；
    所述红外散斑中心亮度模块用于将所述整体亮度平均值与预设区域亮度阈值进行比较，当所述整体亮度平均值与预设区域亮度阈值相一致时，判定所述中心区域实际亮度与预设亮度相一致。
PCT/CN2019/070760 2018-01-10 2019-01-08 体感摄像头的成像精度的调节方法及调节装置 WO2019137348A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US16/641,653 US10986289B2 (en) 2018-01-10 2019-01-08 Method and device for regulating imaging accuracy of motion-sensing camera

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810024754.9 2018-01-10
CN201810024754.9A CN108234874B (zh) 2018-01-10 2018-01-10 体感摄像头的成像精度的调节方法及调节装置

Publications (1)

Publication Number Publication Date
WO2019137348A1 true WO2019137348A1 (zh) 2019-07-18

Family

ID=62640201

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/070760 WO2019137348A1 (zh) 2018-01-10 2019-01-08 体感摄像头的成像精度的调节方法及调节装置

Country Status (3)

Country Link
US (1) US10986289B2 (zh)
CN (1) CN108234874B (zh)
WO (1) WO2019137348A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113936315A (zh) * 2021-10-14 2022-01-14 北京的卢深视科技有限公司 Doe脱落检测方法、装置、电子设备和存储介质

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108234874B (zh) 2018-01-10 2020-07-21 南京华捷艾米软件科技有限公司 体感摄像头的成像精度的调节方法及调节装置
CN111885311B (zh) * 2020-03-27 2022-01-21 东莞埃科思科技有限公司 红外摄像头曝光调节的方法、装置、电子设备及存储介质
CN113240630B (zh) * 2021-04-16 2022-07-12 深圳市安思疆科技有限公司 散斑图像质量评估方法、装置、终端设备及可读存储介质
CN113762253B (zh) * 2021-08-24 2022-08-26 合肥的卢深视科技有限公司 散斑的提取方法、装置、电子设备及存储介质
CN115994904B (zh) * 2023-03-22 2023-05-30 山东万重山电子有限公司 基于计算机视觉的挂烫机面板生产质量检测方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102622591A (zh) * 2012-01-12 2012-08-01 北京理工大学 3d人体姿态捕捉模仿系统
US20130107002A1 (en) * 2011-10-26 2013-05-02 Olympus Corporation Imaging apparatus
CN106412426A (zh) * 2016-09-24 2017-02-15 上海大学 全聚焦摄影装置及方法
CN106502379A (zh) * 2016-09-12 2017-03-15 深圳奥比中光科技有限公司 一种交互方法及交互系统、相对深度的获取方法
CN108234874A (zh) * 2018-01-10 2018-06-29 南京华捷艾米软件科技有限公司 体感摄像头的成像精度的调节方法及调节装置

Family Cites Families (27)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2007072044A1 (en) * 2005-12-23 2007-06-28 Ingenia Holdings (Uk) Limited Optical authentication
US8224018B2 (en) * 2006-01-23 2012-07-17 Digimarc Corporation Sensing data from physical objects
WO2008079301A2 (en) * 2006-12-21 2008-07-03 Massachusetts Institute Of Technology Methods and apparatus for 3d surface imaging using active wave-front sampling
CN102540638B (zh) * 2012-01-05 2015-04-01 中航华东光电有限公司 一种焦点位置检测装置及其检测方法
CN102710951B (zh) * 2012-05-09 2014-06-25 天津大学 基于散斑结构光深度相机的多视点计算成像方法
US9456744B2 (en) * 2012-05-11 2016-10-04 Digilens, Inc. Apparatus for eye tracking
WO2013185936A1 (en) * 2012-06-13 2013-12-19 Koninklijke Philips N.V. Apparatus and method for estimating a property of a surface using speckle imaging
CN103501406B (zh) * 2013-09-16 2017-04-12 北京智谷睿拓技术服务有限公司 图像采集系统及图像采集方法
WO2015077455A1 (en) * 2013-11-25 2015-05-28 Digimarc Corporation Methods and systems for contextually processing imagery
CN103796004B (zh) * 2014-02-13 2015-09-30 西安交通大学 一种主动结构光的双目深度感知方法
JP6361927B2 (ja) * 2014-08-13 2018-07-25 ソニー株式会社 情報処理装置、情報処理方法、およびプログラム
CN204615932U (zh) * 2015-05-21 2015-09-02 广州市艾乐特电子科技有限公司 一种红外变焦一体摄像机
US10749312B2 (en) * 2015-05-28 2020-08-18 Vixar, Inc. VCSELs and VCSEL arrays designed for improved performance as illumination sources and sensors
US20170032527A1 (en) * 2015-07-31 2017-02-02 Iwk Health Centre Method and system for head digitization and co-registration of medical imaging data
CN105095894A (zh) * 2015-08-06 2015-11-25 磐纹科技（上海）有限公司 非接触型图书扫描设备
CN105866132B (zh) * 2016-05-27 2018-08-24 中国铁道科学研究院 一种车载信号机外观检测系统及方法
CN106303174B (zh) * 2016-08-12 2019-08-13 中国科学院光电技术研究所 一种光场相机中微透镜阵列倾斜的数字化调整方法
US9959612B2 (en) * 2016-08-31 2018-05-01 The United States Of America, As Represented By The Secretary Of The Navy Measuring optical turbulence using cell counting algorithms
CN106412433B (zh) * 2016-10-09 2019-01-29 深圳奥比中光科技有限公司 基于rgb-ir深度相机的自动对焦方法及系统
CN106767529B (zh) * 2016-12-14 2019-11-05 深圳奥比中光科技有限公司 激光光斑识别及激光投影仪的自动调焦方法与系统
CN106707526A (zh) * 2016-12-14 2017-05-24 深圳奥比中光科技有限公司 光源投影仪的自动调焦方法与系统
GB2561537B (en) * 2017-02-27 2022-10-12 Emteq Ltd Optical expression detection
US10152798B2 (en) * 2017-04-10 2018-12-11 Wisconsin Alumni Research Foundation Systems, methods and, media for determining object motion in three dimensions using speckle images
CN107300722A (zh) * 2017-07-18 2017-10-27 孙天宇 一种机场跑道异物检测系统
CN107517346B (zh) * 2017-07-31 2020-08-07 Oppo广东移动通信有限公司 基于结构光的拍照方法、装置及移动设备
WO2019027506A1 (en) * 2017-08-01 2019-02-07 Apple Inc. DETERMINATION OF CLEAR-PATTERN LIGHTING VERSUS DENSE
CN108648225B (zh) * 2018-03-31 2022-08-02 奥比中光科技集团股份有限公司 目标图像获取系统与方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130107002A1 (en) * 2011-10-26 2013-05-02 Olympus Corporation Imaging apparatus
CN102622591A (zh) * 2012-01-12 2012-08-01 北京理工大学 3d人体姿态捕捉模仿系统
CN106502379A (zh) * 2016-09-12 2017-03-15 深圳奥比中光科技有限公司 一种交互方法及交互系统、相对深度的获取方法
CN106412426A (zh) * 2016-09-24 2017-02-15 上海大学 全聚焦摄影装置及方法
CN108234874A (zh) * 2018-01-10 2018-06-29 南京华捷艾米软件科技有限公司 体感摄像头的成像精度的调节方法及调节装置

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113936315A (zh) * 2021-10-14 2022-01-14 北京的卢深视科技有限公司 Doe脱落检测方法、装置、电子设备和存储介质
CN113936315B (zh) * 2021-10-14 2022-03-25 北京的卢深视科技有限公司 Doe脱落检测方法、装置、电子设备和存储介质

Also Published As

Publication number Publication date
CN108234874B (zh) 2020-07-21
US20210006734A1 (en) 2021-01-07
CN108234874A (zh) 2018-06-29
US10986289B2 (en) 2021-04-20

Similar Documents

Publication Publication Date Title
WO2019137348A1 (zh) 体感摄像头的成像精度的调节方法及调节装置
US20230293007A1 (en) Method of identifying iris
CN105335722B (zh) 一种基于深度图像信息的检测系统及方法
US9008375B2 (en) Security improvements for iris recognition systems
US20050111705A1 (en) Passive stereo sensing for 3D facial shape biometrics
EP3483832B1 (en) Image processing device
KR101444538B1 (ko) 3차원 얼굴 인식 시스템 및 그의 얼굴 인식 방법
JP6535223B2 (ja) 瞬目計測方法、瞬目計測装置、及び瞬目計測プログラム
JP2007011667A (ja) 虹彩認証装置および虹彩認証方法
JP2013033206A (ja) 投影型表示装置、情報処理装置、投影型表示システム、およびプログラム
JP6651911B2 (ja) 顔画像処理装置
US11501517B2 (en) Individual identifying device
CN113132613A (zh) 一种摄像头补光装置、电子设备和补光方法
JP7159242B2 (ja) 顔生体の検出方法及び検出装置、電子機器、並びにコンピュータ読み取り可能な媒体
JP4968922B2 (ja) 機器制御装置及び制御方法
WO2016172856A1 (zh) 用于识别人脸的装置、系统、方法和计算机程序产品
TW202004660A (zh) 基於前景影像的背景虛化方法與電子裝置
JP2019117579A5 (zh)
KR20120123525A (ko) 이물 판정 장치, 이물 판정 방법 및 이물 판정 프로그램
WO2019163066A1 (ja) なりすまし検知装置、なりすまし検知方法、及びコンピュータ読み取り可能な記録媒体
US11163994B2 (en) Method and device for determining iris recognition image, terminal apparatus, and storage medium
JP2014137762A (ja) 物体検出装置
JP6904624B1 (ja) クラブヘッド計測用マークおよび画像処理装置
JP2002000567A (ja) 瞳孔中心位置計測方法及び視点位置検出方法
US20220309831A1 (en) Techniques for detecting a three-dimensional face in facial recognition

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: 19738303

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: 19738303

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 10/03/2021)