2D/3D real-time imager and corresponding imaging methods
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
H04N-015/00
H04N-013/04
H04N-009/47
H04N-013/02
G01S-007/483
G01S-017/89
출원번호
US-0995697
(2011-12-21)
등록번호
US-9392259
(2016-07-12)
우선권정보
EP-10196702 (2010-12-23)
국제출원번호
PCT/EP2011/073687
(2011-12-21)
§371/§102 date
20130725
(20130725)
국제공개번호
WO2012/085151
(2012-06-28)
발명자
/ 주소
Borowski, André
출원인 / 주소
FASTREE3D S.A.
대리인 / 주소
Young & Thompson
인용정보
피인용 횟수 :
5인용 특허 :
3
초록▼
The present invention relates generally to methods and devices of generating an electrical representation of at least one object in a scene in the real word. The detail real-time imager for the representation of a scene of a real world comprises:—at least an illuminator (0501-0511) of said scene pro
The present invention relates generally to methods and devices of generating an electrical representation of at least one object in a scene in the real word. The detail real-time imager for the representation of a scene of a real world comprises:—at least an illuminator (0501-0511) of said scene providing at least a series of ultra-short power laser pulses with time-related positions; and—a receiver (0515-0523) of a SPAD Single Photon Avalanche Diode detector array according to the method of the invention and associated to at least said series of ultra-short power laser pulses of said illuminator.
대표청구항▼
1. A real-time imager that provides at least one of 3D and 2D detail, comprising: an illuminator comprising:means for generating at least one series of ultra-short laser pulses generated with at least one laser source emitting at least one given wavelength,a receiver comprising:optical and mechanica
1. A real-time imager that provides at least one of 3D and 2D detail, comprising: an illuminator comprising:means for generating at least one series of ultra-short laser pulses generated with at least one laser source emitting at least one given wavelength,a receiver comprising:optical and mechanical means for receiving a flow of light from the illuminated scene when the said illuminator is operative; andmeans coupled to the said optical and mechanical means for receiving a flow of light and for generating a signal representing the targeted scene on the basis of the reflected light from the scene impinging at least one single-photon avalanche diode (SPAD) detector array; said means generating a signal representing the targeted scene and being coupled to an imaging part under at least one programmed controller,wherein the ultra short laser pulses are generated through an electronically controlled modulator to form packets of pulses into the said series of pulses, under control of a means for controlling said means for generating at least a series of ultra short laser pulses; and wherein the illuminator comprises optical and mechanical means coupled to said means for generating at least one series of ultra-short laser pulses for producing at least a predetermined spatial pattern of illumination of a scene onto at least a restricted area in at least one of in azimuth and in elevation,wherein said laser source emitting at a given wavelength comprises a mode-locked laser source operating in the infrared spectral region to provide ultra-short pulses, preferably with a duration of the order of a picosecond, and at a low average power level, optionally coupled to an intermediate semiconductor optical amplifier to amplify low power pulses to a medium power level, then to an rare-earth-doped, preferably erbium, amplifier to amplify the medium power to a high peak power. 2. The real-time imager according to claim 1, further comprising a wavelength combiner which collects the emission from various sources of illumination comprising at least: infrared for 3D, infrared for video, visible light or any other source providing light adapted to the said receiver. 3. The real-time imager according to claim 1, wherein the optical and mechanical means for generating an illuminating pattern comprises a viewfinder which comprises: a movable mirror; andfast variable focusing optics;which are both electronically controlled according to predetermined procedures involving the illuminator and the receiver of the imager. 4. The real-time imager according to claim 3, wherein the viewfinder comprise also a light homogenizer receiving the light from the amplifiers, and the optics are adjusting the size of an illuminating beam of a selected section, square or round, to the target distance. 5. The real-time imager according to claim 3, wherein the viewfinders of both the illuminator and the receiver are enclosed in the same or different casing with a transparent opening. 6. The real-time imager according to claim 3, wherein each optical and mechanical part requiring a bearing function cooperates preferably with at least an air bearing with air supplying through the center of axis. 7. The real-time imager according to claim 6, wherein said air bearing comprises a plurality of jets with at least one bush which moves up and down relatively to the center axis due to the pressure onto the jets and dispositive to save air leak and require no moving cables. 8. The real-time imager according to claim 6, wherein a optical and mechanical means comprises a variable zoom and a rotating mirror which use at least one of the group consisting of: a servo-motor controlling the horizontal azimuth of any mirror;air bearing supporting any mirror in all its quick movements;an ultra light mirror aimed in azimuth and elevation to the target point;a tilting electro-pneumatic actuator driving the mirror at the chosen elevation with strong damping capabilities;a top motor and bottom motor controlling the movement and tension of a cable attached to the lenses of the zoom; andinterferometers controlling the movement of any moving part of the viewfinders and connected to respective controllers. 9. The real-time imager according to claim 8, further comprising an electric motor, that is electrically powered using wireless transmission or a rotating transformer; a wear-free contact between the air bearing shaft and stationary parts is provided preferably by means of a capacitive coupling, which is used to transfer information of the at least one of the speed of rotation and the angular position of the axis of the mirror to a rotating electronic control system. 10. The real-time imager according to claim 8, further comprising some variations in air pressure that is being fed down the center of the shaft, the mirror axis air bearing operating over a wide range of supply pressure such that a variation of this pressure controls the position of the mirror. 11. The real-time imager according to claim 8, wherein zoom optics are aligned onto a central axis of the viewfinder, a plate carrying some optional filters is interposed at the output of the zoom along the optical path, a motor allowing selection of the convenient filter under control of a dedicated controller, a mirror rotating box is aligned onto the central axis and contains a dichroic mirror and at least a CCD sensor, which is a 2D image sensor operating in a visible range of light, and at least a 3D infrared sensor, preferably a single-photon avalanche diode (SPAD) detector, and a motor to rotate the mirror rotating box to rapidly connect the designated CCD sensor at the control of a dedicated controller. 12. The real-time imager according to claim 1, wherein optical and mechanical means for receiving a flow of light at the receiver comprises a viewfinder with optics having a tilting and rotating mirror collecting the faint light reflected from the target and directing it to a zoom mechanism that will adjust the focus to the distance of the target. 13. The real-time imager according to claim 1, further comprising CCD sensors to receive several wavelengths for 2D data from the target through common optics together with the 3D data acquired in the infrared range with said single-photon avalanche diode (SPAD) detector array, a dichroic mirror to separate the wavelengths such that the 3D data capture is always available and the 2D acquisition run at successive phases, one for each wavelength to improve the sensitivity, some rotating filter wheels for selecting different wavelengths and variable focusing mechanisms. 14. The real-time imager according to claim 1, wherein an alternative sensor is switched in by a monogon rotating dichroic mirror in the optical path, thus rotating the target image on 2D sensors, this rotation being corrected in software to obviate some aberrations. 15. The real-time imager according to claim 1, further comprising at least a real-time electronic board provided to support various controllers to manage the following subsystems: modulator, light amplifier, zoom, viewfinder, single-photon avalanche diode (SPAD) matrix, DSP, power management, internal data transfers, and temperature calibration. 16. The real-time imager according to claim 15, wherein an independent board comprises a means for controlling safety of the illuminator. 17. The real-time imager according to claim 15, wherein an independent board comprises means for managing security, local and remote access and external data transfer to a central controller through an Ethernet-like link. 18. The real-time imager according to claim 15, wherein a real-time electronic board comprises means for managing the tactical acquisition activity of the imager. 19. The real-time imager according to claim 18, wherein the real-time electronic board comprises means for managing the maximum power/precision of each frame by changing the operational parameters of the modulator and the semiconductor amplifier. 20. The real-time imager according to claim 18, wherein the real time electronic board comprises means for operating the imager in different capture modes, provided by a capture mode selector between: 1) a Landscape acquisition (light landscape, plain landscape imagers);2) crude detail acquisition;3) full detail acquisition;4) large area, detailed acquisition. 21. A real-time imager that provides at least one of 3D and 2D detail, comprising: an illuminator comprising:means for generating at least one series of ultra-short laser pulses generated with at least one laser source emitting at least one given wavelength,a receiver comprising:optical and mechanical means for receiving a flow of light from the illuminated scene when the said illuminator is operative; andmeans coupled to the said optical and mechanical means for receiving a flow of light and for generating a signal representing the targeted scene on the basis of the reflected light from the scene impinging at least one single-photon avalanche diode (SPAD) detector array; said means generating a signal representing the targeted scene and being coupled to an imaging part under at least one programmed controller,wherein the ultra short laser pulses are generated through an electronically controlled modulator to form packets of pulses into the said series of pulses, under control of a means for controlling said means for generating at least a series of ultra short laser pulses; and wherein the illuminator comprises optical and mechanical means coupled to said means for generating at least one series of ultra-short laser pulses for producing at least a predetermined spatial pattern of illumination of a scene onto at least a restricted area in at least azimuth or in elevation,wherein the single-photon avalanche diode (SPAD) detector array of the receiver is built on an integrated circuit which embeds:the single-photon avalanche diode (SPAD) detector array;processing means which apply one or more treatments to the detected signal at the single-photon avalanche diode (SPAD) detector array itself, to ensure at least a high image resolution or the capability to represent 3D movements of objects of a real scene;digital signal processors distributed from the cell level built around a single single-photon avalanche diode (SPAD) with a local DSP, a group of neighbouring single-photon avalanche diode (SPAD) cells clustered in a macrocell having its proper DSP, to a global level DSP for the overall of the treatments generated at the local DSPs. 22. The real-time imager according to claim 21, wherein optical and mechanical means for receiving a flow of light at the receiver comprises a viewfinder with optics having a tilting and rotating mirror collecting the faint light reflected from the target and directing it to a zoom mechanism that will adjust the focus to the distance of the target. 23. The real-time imager according to claim 21, wherein the viewfinders of both the illuminator and the receiver are enclosed in the same or different casing with a transparent opening. 24. A method for acquiring 3D scenes from such at least one of a 3D detail real-time imager 3D and 2D detail real-time imager, comprising: an illuminator comprising:means for generating at least one series of ultra-short laser pulses generated with at least one laser source emitting at least one given wavelength,a receiver comprising:optical and mechanical means for receiving a flow of light from the illuminated scene when the said illuminator is operative; andmeans coupled to the said optical and mechanical means for receiving a flow of light and for generating a signal representing the targeted scene on the basis of the reflected light from the scene impinging at least one single-photon avalanche diode (SPAD) detector array; said means generating a signal representing the targeted scene and being coupled to an imaging part under at least one programmed controller,the ultra short laser pulses being generated through an electronically controlled modulator to form packets of pulses into the said series of pulses, under control of a means for controlling said means for generating at least a series of ultra short laser pulses; and the illuminator comprising optical and mechanical means coupled to said means for generating at least one series of ultra-short laser pulses for producing at least a predetermined spatial pattern of illumination of a scene onto at least a restricted area in at least azimuth or in elevation, wherein the method comprises steps of:generating time series of at least a wavelength of a time series of ultra-short power laser pulses;forming beams of ultra short power laser pulses generated at different angles for illuminating a restricted area of a scene to be imaged, both in azimuth and elevation;receiving reflected laser pulses onto at least a SPA single-photon avalanche diode (SPAD) detector array and deriving 3D coordinates of each illuminated dot of the illuminated scene from the knowing of each photon impinging a given single-photon avalanche diode (SPAD) cell of the single-photon avalanche diode (SPAD) detector array, and of the related angle of illuminated dot of the scene, the data being locally processed such that they are blanked filtered, averaged at least one of in time and in neighboring single-photon avalanche diode (SPAD) cells clustered in macro-cells, and at least one of compressed and transmitted to an imaging part. 25. The method according to claim 24, further comprising a step of selecting the number of generated pulses in each of a plurality of packets of the said time series of ultra-short power laser pulses according to a selected strategy of illuminating the scene. 26. The method according to claim 24, further comprising a step of selecting at least one of a laser and a light source of a given wavelength to illuminate the scene. 27. The method according to claim 24, further comprising a selected rotation of mirrors of at least one of said viewfinders and the state of controllable zooms of the said viewfinders. 28. The method according to claim 24, wherein in a first step, a general view of the scene is acquired at a given resolution on a 3D basis, for acquiring some interesting targets, then in a second step, a detailed acquisition occurs based on the exact knowledge of the expected distance of the targets acquired or analyzed at at least one of the end of the said first step and on the basis of other means of knowing the target distance could be used like optical settings or knowledge by other means. 29. The method according to claim 28, wherein a first part of the second detailed acquisition step is a verification frame without blanking to correct the possible movement of the target and precision inaccuracies between the landscape and detail acquisitions. 30. The method according to claim 29, wherein at a second part of the second detailed acquisition step, a blanking signal is generated toward the single-photon avalanche diode (SPAD) detector array which is enabled so that only packet of received pulses corresponding to an approximate position of each illuminated dot of the target is precisely acquired, said blanking signal being generated on the basis of the coarse acquisition at the first step of acquisition of the 3D landscape scene and applied to an enabling circuit associated with a single-photon avalanche diode (SPAD) cell so that the received photons when the blanking signal is high are not processed. 31. The method according to claim 24, wherein the emitted power level is constantly optimized on the basis of an automatic gain control mechanism. 32. The method according to claim 24, wherein the size and the resolution of the macro-cells of the single-photon avalanche diode (SPAD) detector array is changeable. 33. The method according to claim 24, wherein a step of detecting the intrusion of a foreign body is performed at high speed using optimal energy and time mainly at the edges of the said restricted area of illuminating. 34. The method Method according to claim 24, wherein a 3D acquisition process is supplemented or substituted by a simultaneous 2D acquisition process in at least one of the IR and visible ranges. 35. The method according to claim 24, further comprising a selection between three or more image qualities comprising: a full quality images;a good enough quality images; anda fast movement capture. 36. The method according to claim 24, further comprising at least one of the following steps: collimating laser pulse beams on variable angular size targets;collecting the imaged reflections of those pulses to the matrix sensors;using preferentially separate optics for both tasks;providing a large dynamic of measurement distances while keeping optical quality high and limiting the number of moving block;using periodic calibration of the optical settings on fixed external elements instead of perfect optic to save cost and space;performing a 360° horizontal angular range of measurement;performing +/−15° of tilting on the elevation;acquiring several positions of measurement by second;controlling a continuously and smoothly scanning of the 360° angular range. 37. The method according to claim 24, wherein, to ensure safety of living target, the method further comprises at least one or more of the following steps: selecting a wavelength in the 1.5 μm IR wavelength, providing an inherent eye safety;selecting the shortest possible lowest energy measurement packets of ultra-short power laser pulses generated at the illuminator;selecting limitations on the number of measurement packets of ultra-short power laser pulses generated at the illuminator in a second of time in a given angular range only, wherein this number will change with the emitted power and then with the distance;adapting the maximum power emitted to the size of the surface area illuminated;real-time supervising the emitted power by an independent subsystem inside the imager, wherein this supervision subsystem will control the emitted power and also the movement of the mirrors in charge of the angular distribution;self checking all security systems to be active permanently;managing the whole transmitted energy, in all wavelengths, to the target in the same second by an individual imager or a group of imager coordinated explicitly or by using self healing optimization;preferentially using a landscape-mode measurement before any detail measurement to minimize the emitted power during a detail measurement;using a periodic non-blanked pulse measurement to detect an intrusion in the measurement field, from the sides, wherein an intrusion is detected by the presence of any target 90% farther than a few meters from the expected target distance;reducing all beams power in a limited angular range of 10° angular range direction when an intrusion is detected; anddetecting a possible proximity presence by using a proximity detector if the other measures are not efficient to detect a safety problem in this distance range. 38. The method according to claim 24, wherein a frame for an image acquisition is controlled with one or more of the following steps: resetting the various parts of the imager;selecting at least a laser pulse by the modulator according to a given strategy ordered to build convenient time series of ultra-short power laser pulses;amplifying the selected pulse by the semiconductor amplifier;amplifying said amplified laser pulse is also by the Erbium amplifier;beaming said ultra-short power laser pulse by the optics and the mirror of the illuminator of the imager;reflecting said laser pulse by the target;capturing said reflected beam by the optics and mirror of the viewfinder dedicated to the receiver of the 3D detail real-time imager;avalanching photons of said reflected beam onto at least one single-photon avalanche diode (SPAD) detector, a diode which is addressed onto the matrix form of the single-photon avalanche diode (SPAD) detector matrix, to get a timestamp related to the timestamp of the emitted laser pulse;filtering, blanking, averaging at least one of in time and in space, neighbouring single-photon avalanche diode (SPAD) detectors, preferably clustered in at least one of in macro-cells and locally compressing; andcompressing at a global level DSP of the data after they have been at least one of averaged and locally compressed and transferring it to the external world. 39. The method according to claim 24, wherein several 3D detail real-time imagers are clustered towards a same scene to compose complex images of scenes, wherein the acquisitions of images are sequenced such that data can be shared between the various imagers; each imager is calibrating that imager's scene acquisition in such way that a first frame is acquired at a first low resolution and details of the acquired frame and then subsequent frames are acquired at at least one of other resolutions and other positions of the acquired zone in the scene.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (3)
Kim, Seong Jin, Apparatus and method of obtaining depth image.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.