3D imaging system and 3D imaging method
The 3D imaging system encodes range information as color using wavelength modulation, addressing the precision timing and pixel area challenges of existing LIDAR systems, achieving efficient and cost-effective 3D imaging.
Patent Information
- Application Number
- CN201910218606.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2018-03-21
- Filing Date
- 2019-03-21
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2039-03-21
AI Technical Summary
Existing 3D flash LIDAR systems require high precision timing circuits and large pixel areas, leading to reduced sensitivity and imaging resolution.
A 3D imaging system using an optical modulator to differentially amplitude modulate two wavelengths in reflected light, encoding range information as color in the image, eliminating the need for complex timing circuits and allowing the use of off-the-shelf detector arrays.
Enables high-resolution 3D imaging without precise timing circuits, reducing complexity and cost while maintaining sensitivity and resolution.
Smart Images

Figure CN110297255B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a system for 3D imaging and a system for determining the range of at least one object within a field of view, in particular for use as part of an autonomous or semi-autonomous vehicle. The invention also relates to methods of 3D imaging and range determination. Background Art
[0002] The use of 3D imaging devices to capture the 3D position of objects in a scene is becoming increasingly common for applications such as autonomous vehicles. By detecting the position, and in particular the range, of objects in the scene around the vehicle, the speed and direction of the vehicle can be controlled to ensure that the vehicle does not collide with any objects, or to enable the vehicle to track objects moving in the scene. In order to detect small objects at large distances, the imaging device must have a high resolution. The high resolution image can be fed to a suitable signal processing unit that forms part of the vehicle control system.
[0003] One specific technology that can be used to capture 3D images of a scene is called LIDAR. The principle behind LIDAR devices is to bounce pulses of light off objects to see where the object is, just as radar uses radio waves and sonar uses sound. LIDAR systems send out pulses of light outside the visible spectrum, although the light can be within the visible spectrum in some applications, and how long they take to bounce back. The time it takes for the light to travel from the source, reflect off the object, and bounce back to the detector tells the sensor how far away the object is.
[0004] LIDAR systems can generate a large number of narrow, focused light pulses, each of which is directed to a small spot in the scene. The spot is scanned in two dimensions, and the resulting set of reflection timings can then be plotted on a 3D map called a "point cloud" to build up a detailed picture of the surrounding environment. This requires only a single element detector, but a complex set of scanning equipment must be provided to build the point cloud map. In addition, the detector requires high-precision timing circuitry to measure the travel time of the light.
[0005] Another LIDAR technology is called 3-D flash LIDAR. This has an advantage over other scanning LIDAR technologies because it avoids the need for a scanner to physically scan a light source across the scene. In a typical 3-D flash LIDAR system, the detector consists of a 2D array of pixels, each of which records the time it takes for a short-duration, wide-angle laser flash pulse to travel into the scene and bounce off all objects in the scene to reach the image detector. Each detector element or pixel / pixel group has an independent trigger and counter to record the flight time of the laser pulse to the object. The physical range of the object in front of the camera is calculated based on the flight time information.
[0006] The main problem with time-of-flight based 3-D flash laser LIDAR is the very precise trigger timing required to measure the flight time of the laser pulse with sufficient accuracy to achieve the desired range resolution. A second problem is the large amount of pixel area required per pixel electronics: thus, reducing sensitivity and / or requiring an increase in pixel size (and a corresponding reduction in imaging resolution). Summary of the invention
[0007] It is an object of the present invention to provide an improved system and apparatus for obtaining 3D images which does not require high precision timing circuits for the pixels of the detector.
[0008] According to a first aspect, the present invention provides a 3D imaging system, comprising:
[0009] a light source configured to illuminate a scene with at least one light pulse, the light pulse comprising light having at least two discrete wavelengths;
[0010] an optical detector comprising an array of detection elements; and
[0011] An optical modulator is located in an optical path between the scene and the optical detector, the modulator being arranged to differentially amplitude modulate two wavelengths in a reflected portion of the light pulse as a function of time.
[0012] In use, the 3D imaging system of the present invention can produce a 2D image of a scene in an XY reference frame, wherein each point is assigned distance or range information to an object imaged at that point. In use, it is also within the scope of one aspect of the present invention for the system to identify the location of one or more objects in the XY frame and the range of the objects in the scene without producing a complete image of the scene, or simply outputting range information. For example, it can output a signal indicating the range of the nearest object in the scene in the simplest arrangement, or even output an alarm if the range of the nearest image is below a minimum threshold range.
[0013] The system may include focusing optics that focus a portion of a light pulse that has been reflected or scattered from an object in the scene onto an optical detector.
[0014] The light source may be arranged to illuminate the scene with at least two sets of temporally spaced light pulses, each set of pulses comprising at least one light pulse having at least two discrete wavelengths. Each set may cause reflected light to impinge on the detector in an order that allows the detector to collect light from each pulse before the integrator reads out a value forming an image. Thus, two different images may be output from the detector based on the two sets.
[0015] In use, the system may transmit groups of pulses continuously, or in batches at defined times when a 3D image of a scene is required.
[0016] The optical detector may include a single array of detector elements, typically arranged in a regular two-dimensional grid, wherein each detector element in the detector array responds to two wavelengths to produce an output signal having a value indicative of the spectral content or color of light incident on the detector element. In this way, the detector array will produce a corresponding color image from the set of output signals for each of the two groups of pulses according to the values output by the array of detector elements. Notably, the detector array does not require any of the complex and high precision timing circuitry of prior art LIDAR systems because the range is encoded as the color in the image rather than the time at which the light arrives at the detector.
[0017] Suitable detectors that may be used in the systems of the present invention include Bayer filter array sensors.
[0018] Each detector element may include at least two pixels, each pixel responsive to a different one of the two wavelengths, and the output of the element may include an intensity value from each pixel or a single spectral component or color value obtained by processing the values from the two pixels together. Suitable pixelated detectors for various cameras for both the visible and invisible spectrum are known to those skilled in the art, and because the individual pixels do not require the complex trigger electronics of prior art LIDAR systems, the complexity and cost of the detector may be minimized by using widely available off-the-shelf components.
[0019] In the context of the present invention, the term "color" refers to the spectral composition of the light incident on each portion of the detector. Where the detector comprises an array of detector elements, the term "color" refers to the spectral composition of the light incident on each detector element. It is not limited to the visible spectrum of colors.
[0020] By modulating the two wavelengths differentially in time, light impinging on the detection element of the detector from a close object will include a different ratio of the first wavelength to the second wavelength than light impinging on the detection element subsequently from a more distant object. This is because the round trip time from source to detector is shorter for a close object than for a more distant object, and because the modulation applied at these two times is different.
[0021] The modulator may modulate the reflected light over a period of time equal to or greater than the time it takes for light from a single pulse to travel from the source to the farthest object in the scene and back to the detector.
[0022] The modulation period of the modulator may be synchronized with the emission of the light pulses so that each modulation period begins when the emission pulse is generated by the source or at some known time delay after the start. In any case, it should begin before light from the nearest possible object in the scene is reflected or scattered back to the detector and should not end before light from the farthest object that one wishes to detect has been reflected back to the detector.
[0023] In order to synchronize the modulation period with the light pulses, a control unit may be provided which generates a common trigger signal for the light source and the modulator. Each time the trigger signal is generated, a pulse may be emitted and a modulation period may be started.
[0024] A set of pulses may include 1, 2 or more than 2 pulses, each pulse being identical in spectral composition over the pulse duration. Each pulse may include a broadband pulse of white light from a white light source, or may include two different narrow wavelength bands. The two narrow bands may be generated by corresponding lasers, each laser having a different wavelength.
[0025] Thus, each pulse may be generated by a single source emitting at two or more wavelengths, or from multiple sources emitting light simultaneously and each emitting at one or more of the two wavelengths.
[0026] To direct the reflected or scattered light onto the detector, the system may include focusing optics upstream of the modulator, such as a single focusing lens, which forms an image of the scene in a focal plane. The detector may be located in the focal plane. The design of suitable focusing optics is familiar to those skilled in the art of camera system design and is therefore not described in detail here.
[0027] The function of the optical modulator is to differentially amplitude modulate the spectral content of light passing through or reflected from it as a function of time to encode the reflected light with range information.
[0028] Although there may be two wavelengths that are modulated, the invention can be applied to broadband light sources with more than two wavelengths, where the spectral content is differentially amplitude modulated at more than two wavelengths. For example, white light can be used, and the modulator can attenuate the light to apply a continuously or step-wise variable tint from one end of the spectrum to the other. In the case of white light in the visible spectrum, the tint can be varied through the spectrum from the red end to the blue end of the spectrum in discrete steps or in a continuously variable manner.
[0029] The modulator can be amplitude modulated in the form of a notch filter with a varying center wavelength or in the form of a bandpass filter with a varying center wavelength or in a different manner.
[0030] The modulator can operate in more than one mode.
[0031] One suitable optical modulator comprises a variable wavelength filter that differentially modulates two wavelengths in a first operating mode such that a first wavelength is incrementally attenuated over time during a modulation period and a second wavelength is either not attenuated or is incrementally attenuated, or is attenuated by a fixed amount, over that time.
[0032] The optical modulator can also operate in a second mode, whereby it modulates two wavelengths, whereby the second wavelength decays incrementally over time, and the first wavelength either does not decay, decays incrementally, or decays by a fixed amount.
[0033] In each mode, the modulation can vary linearly over the entire modulation period, or it can vary nonlinearly. It can vary in a stepwise manner or continuously.
[0034] Providing a modulator that can operate in two modes enables range information to be obtained without any compensation for the effect of the colour of objects on the scene, which themselves will modulate the light from the light source.
[0035] To exploit this feature, in use, the detector may be arranged to capture a first image of the scene corresponding to a first set of pulses, whereby for each pulse the modulator operates only in the first mode, and wherein the detector captures a second image of the scene corresponding to a second set of pulses, whereby for each pulse the modulator operates only in the second mode.
[0036] The first image is captured at a different time than the second image, allowing a detector array to sequentially form the first and second images. The apparatus may include signal processing circuitry that determines the extent of an object in the scene by comparing the spectral content of light encoded in the color values for each detector element in the array in the first image with the spectral content of light encoded in the color values for each corresponding detector element in the array in the second image.
[0037] A point in a 3D image can be generated by comparing the color or intensity (depending on the specific system configuration) of the light encoded in the value of the signal output by the detector element across the detector element. For example, using a lookup table that relates the resulting "comparison value" to a corresponding range value. The "relative intensity" value RI can be determined using the following equation:
[0038] RI = [(I(B)%*object reflectivity (B)) + (I(R)%*object reflectivity (R))]*[inverse square range factor]*[modulation factor]
[0039] Where the first and second wavelength intensities I(B) and I(R) depend on the spectral composition of the emitted light and the color of the reflected object, the electro-optic modulator goes from 0% first wavelength to 100% first wavelength (and vice versa for the second wavelength) during the modulation period, and the calculation uses a gray object. Testing using blue and red wavelengths and a gray object at a range of 5 ft as a baseline (at the detector = 100%) produced good results in the experimental model.
[0040] In use, the system of the first aspect may output a 3D image of a scene based on combining the two captured images.
[0041] Applicants propose that a suitable optical modulator for use in practical embodiments of the invention is an electro-optic wavelength modulator which is ramped over a period of time which depends on the maximum distance of the object range to be measured. The effect of the ramping is to change the frequency response of the modulator according to the relative attenuation of the amplitudes of at least two discrete wavelengths in the reflected light.
[0042] The modulator may include a graphene-based electro-optical variable filter modulator. Research into graphene technology has shown that these devices are capable of operating at desired modulation frequencies and low operating voltages, making them suitable for a wide range of applications, such as commercial automotive products. Suitable graphene devices are disclosed in a publication by Erick Ulin-Avila et al., entitled "A Graphene-Based Broadband Optical Modulator"; Nature article, June 2011, DOI: 10.1038 / nature10067, source: PubMed: https:
[0043] / / www.researchgate.net / publication / 51107563_A_graphene-based_broadband_optical_modulator.
[0044] To generate short (~1 ns) white light pulses for transmission, a source such as a white light supercontinuum laser can be used (see, e.g.,
[0045] https: / / www.leica-microsystems.com / science-lab / white-light-laser / ). In contrast, in the case of a dual-wavelength light source, two tunable wavelength or fixed wavelength lasers may be provided.
[0046] Although the two images may be captured in sequence and combined to generate a 3D image, in an alternative arrangement two detector sub-arrays may be provided and the two images captured simultaneously.
[0047] A detector array may be provided comprising a first sub-array of detection elements, each detection element being responsive to a first wavelength and not responsive to a second wavelength, the detector array further comprising a second sub-array of detection elements, each detection element being responsive to a second wavelength and not responsive to the first wavelength, each sub-array being arranged in use to output a monochromatic image.
[0048] A beam splitter may be provided that receives light reflected or scattered from the scene and splits it into two beams, the first beam forming an image of the scene on the first subarray and the second beam forming an image of the scene on the second subarray. The images from the two subarrays may be combined to generate 3D range information.
[0049] The first and second sub-arrays may comprise arrays of colour detectors, each covered with a filter corresponding to a respective one of the two wavelengths, so that each is responsive to only one of the two wavelengths. Alternatively, each may be a monochromatic detector.
[0050] Applicants have also realised that another factor to consider is the potential movement of objects in the field of view between consecutive first and second images. For example, for an image capture rate (frame rate) of the order of 30 fps, this gives a time between consecutive frames of ˜33 ms. The integration time for a single image formed by a set of pulses may be of the order of 1 ms. Many of these effects are common to other types of LIDAR, and some ways of dealing with them in processing the received signal are known.
[0051] The applicant has also realised that the effect of the colour of the object may have a significant impact. It is therefore proposed that the system is configured to eliminate this effect by appropriately processing the colour data from the first and second images.
[0052] Alternatively, the system may be arranged in use to capture a third image when the modulator is operated in a third mode in which the wavelength modulator applies no modulation or applies equal modulation to all wavelengths, the third image providing a reference. The signal processing circuitry may then be arranged to use the third image to compensate for any common mode effects on the first and second images, for example by subtracting the third image from the first and second images or by applying a scaling factor to the first and second images using the third image.
[0053] According to a second aspect, the present invention provides a method for 3D imaging of an object in a scene, comprising:
[0054] illuminating the scene with at least one light pulse;
[0055] differentially amplitude modulating at least two wavelengths in a portion of the pulses reflected or scattered by an object in a scene, and focusing the modulated light onto at least one detector to form an image of the scene; and
[0056] At least one image is processed to determine the extent of objects in the scene.
[0057] The method may include:
[0058] (a) illuminating a scene at a first moment in time with a pulse of light comprising at least two discrete wavelengths, and illuminating the scene at a second moment in time with a second pulse of light having the same two wavelengths;
[0059] (b) directing reflected or scattered light from an object in the scene for the first pulse onto a detector array via an optical modulator arranged to apply a first amplitude modulation to two wavelengths in a reflected portion of the light pulse as a function of time;
[0060] (c) directing reflected or scattered light from an object in the scene for a second pulse onto a detector array via an optical modulator arranged to apply a second different amplitude modulation to two wavelengths in the reflected portion of the light pulse as a function of time; and
[0061] (d) comparing a first image obtained from the detector corresponding to the first pulse with a second image obtained from the detector corresponding to the second pulse to form a 3D image of the scene.
[0062] Step (b) may include modulating the two wavelengths such that the first wavelength decays incrementally over time and the second wavelength does not decay or decays incrementally during the time, and wherein step (c) includes modulating the two wavelengths such that the second wavelength decays incrementally over time and the first wavelength does not decay or decays incrementally.
[0063] The method may be performed using an apparatus having the features according to the first aspect of the invention.
[0064] The method may further comprise capturing a third image when the modulator does not apply modulation to the reflected light.The method may comprise using the third image to compensate for the effect of different colours of objects in the scene causing common mode effects in the first and second images captured by the detector. BRIEF DESCRIPTION OF THE DRAWINGS
[0065] Two embodiments of the invention will now be described by way of example only, in which:
[0066] Figure 1 is an overview of a first embodiment of a 3D imaging system according to an aspect of the present invention;
[0067] Figure 2 (a) shows the spectral response of the modulator over a period of time when operating in a first mode of operation;
[0068] Figure 2(b) shows the spectral response of the modulator over a period of time when operating in the second mode of operation;
[0069] Figure 3 shows the modulation of two wavelengths in light reflected from each object in a scene when the modulator is operated in a first mode for a single light pulse from a light source;
[0070] Figure 4 shows the modulation of two wavelengths in light reflected from each object in the scene when the modulator is operated in the second mode for a single light pulse from the light source;
[0071] Figure 5 showing a first color image captured by the detector for a first set of pulses, wherein a first mode is used for all pulses and objects at different distances are coded with different colors;
[0072] Figure 6 showing a corresponding second color image captured by the detector for a second set of pulses, wherein the second mode is used for all pulses and objects at different distances are coded with different colors;
[0073] Figure 7 is a flow chart showing the steps performed according to the method of the present invention; and
[0074] Figure 8 is an overview of a second embodiment of a 3D imaging system according to an aspect of the present invention. DETAILED DESCRIPTION
[0075] Figure 1 The key components of a first embodiment of a 3D imaging system 100 are shown. The system comprises two main parts; a light source and a color sensing camera. In this example, the light source comprises a white light source 101 that illuminates a scene. The camera is located in a position to image light reflected from objects in the scene.
[0076] The camera includes a detector or color imager 102, which includes: a planar 2D array of detector elements or pixels / pixel groups; a focusing lens 103 that images the scene onto a focal plane containing the 2D array; and an optical modulator 104 located in the optical path between the lens 103 and the detector 102. In one variation, the modulator may be located upstream of the lens or integrated into the lens or detector. The focusing optics may be more complex than a single lens and may include multiple lenses and / or reflectors to control the optical path of light reflected or scattered from objects in the scene,
[0077] The optical modulator 104 in this example comprises a graphene-based electro-optic wavelength modulator. A driver circuit 105 is provided, which applies a variable voltage to the modulator in use. The modulator acts as a transmission filter, wherein the amplitude attenuation of light of different wavelengths passing through the filter can vary over time. Alternatively, the modulator can operate in a reflection mode using a suitable back reflector.
[0078] White light source 101 is driven by driver 106, which causes it to output a series of short light pulses. In this example, each pulse has a duration of approximately 1ns, but other durations may be used within the scope of the present invention. Each light pulse is propagated through a lens in front of the light source to evenly illuminate the scene. Figure 1 Three objects A, B and C are shown in the scene, which are illuminated each time a pulse is generated. A cone 107 in the figure defines the boundary of the scene, any object within the cone is detected by the system, while any object outside is not detected.
[0079] Each light pulse from the light source 101 will be partially reflected or scattered from each of the three objects back to the color sensing camera and in the absence of the modulator 104 a full color image will be formed by the 2D array of detector elements as reflected white light hitting each pixel / pixel group of the array. Figure 1 In the arrangement, before being imaged, the received light signal passes through a graphene-based electro-optical wavelength modulator that encodes differential wavelength amplitude modulation on the white light impinging on the detector. In effect, this applies a time-dependent color tint to the light arriving at the detector with the color of the tint depending on the time at which the light arrives at the detector.
[0080] exist Figure 2 In (a), for simplicity, the wavelength amplitude modulation is shown as ramping linearly from red to blue over a period of time, where the ramp occurs during the time period for the pulse to travel from the laser to the farthest object in the scene and back to the camera. For example, if a maximum range of 50m (object C) needs to be measured, the total travel time of the pulse to object C and back to the detector is (2×50) / 3×108=333ns. Therefore, in this case, consecutive 1ns laser pulses must be transmitted with an intermediate time gap of 333ns.
[0081] exist Figure 2 In (b), opposite ramps are applied and the wavelength modulation is ramped linearly from blue to red over the same period. As will be described, the two ramps are applied to corresponding images in a pair of images and from this the extent of the object can be calculated.
[0082] Figure 3 and Figure 4shows how the light of each object is modulated according to the time it arrives at the modulator. In these figures, the attenuation of red wavelengths is represented by the solid line and the attenuation of blue wavelengths is represented by the dashed line.
[0083] Assume that the wavelength modulation ramp is initially from red to blue, such as Figure 2 (a), white light reaching the camera from a short-distance object such as A will pass through the wavelength modulator while still in the "red" part of its slope, where red wavelengths have little attenuation and blue has high attenuation: the light reaching the array pixels will therefore be predominantly "red". For longer-distance objects, such as B and C, the wavelength modulation slope shifts toward the blue end, with red wavelengths heavily attenuated and blue wavelengths less or not attenuated at all, so the return light reaching the detector pixels becomes increasingly "blue". This is Figure 5 Example in.
[0084] These two images can be used to determine the extent of the object. Figure 7 A suitable exemplary method of using the apparatus is shown in .
[0085] In a first step 200, a first pulse is emitted to objects A, B and C in the scene. At the same time, the driver for the optical modulator triggers Figure 2 (a) shows a modulation ramp 201. The modulated light is focused onto and emitted from the modulator, causing it to strike the array pixels, which record both intensity and color. In practice, many consecutive pulses are emitted sequentially during the imager's integration period (~1 ms), each separated by 333 ns (in this example). For each of these pulses, the wavelength modulator is moved in the same direction (e.g., from red to blue, as in Figure 2 (a)). At the end of the integration period, a color coded image is captured 202 from the array and stored in memory.
[0086] In the next step 203, another set of pulses is output from the light source onto the scene. This time, the driver is triggered synchronously with the pulses so that the modulator applies 204 Figure 2 (b) The modulated reflected light from the set of pulses illuminates the detector array and at the end of the integration period, a second color coded image is generated 205 and stored in memory. Since the wavelength modulation ramp goes from blue to red in opposite directions, nearby objects appear primarily "blue" while more distant objects appear primarily "red", as shown in FIG. Figure 6 shown.
[0087] Each of the two stored images, one for each consecutive pulse train (i.e., each consecutive integration period of the detector), will depend on the direction of “travel” of the electro-optic wavelength modulator (red to blue, or blue to red) to capture the range of different objects represented in the perceived color coding in the image.
[0088] In the next step 206, a pixel-by-pixel comparison of the colors in the two images is performed and the range of each of the three objects visible in both images of the pair is calculated. This is possible as long as the modulation behavior of the modulator is known and can be implemented using a lookup table stored in a memory containing pairs of color values for the first and second image.
[0089] The range information for each pixel is then used to construct a 3D image of the scene, or simply output the spatial position (XYZ) of objects in the scene.
[0090] Alternative configuration
[0091] It is conceivable Figure 1 An alternative configuration of the system shown in . For example, in Figure 8 In an alternative embodiment of the system 300 shown, two laser sources 301 and 302 are provided, each of which simultaneously emits pulses of different wavelengths (shown as red and blue for convenience). At the camera, the returning light signal first passes through a wavelength modulator 301, and then through a beam splitter 302 to two separate detector subarrays 303, 304, which have red and blue filters in front of them respectively. This arrangement eliminates the need to generate very short white light pulses, and also enables the collection of "red" and "blue" images simultaneously (rather than continuously) at the expense of a more complex detector arrangement. In this arrangement, instead of using a color sensitive array, the detector subarrays may include monochrome detectors, which allow the filters to be omitted. In this embodiment, range information is encoded in each of the two images by the relative intensity of objects in the two images.
[0092] In a further modification, the two light sources can be replaced by a single white light source and two detector subarrays are used simultaneously.
[0093] In another alternative configuration, two light sources may be configured with a single color sensing camera that takes sequential images, such as Figure 1 The two light sources together perform the Figure 1 Basically the same function of a single white light source.
[0094] Wavelength selection
[0095] Although the description of the above two examples refers to "white", "red" and "blue" light for ease of explanation, any suitable wavelength may be used. For white light, any broadband light source may be used. In practice, such a system is likely to operate using infrared wavelengths, making the emitted light beam invisible. This is particularly relevant when the system is mounted on a road vehicle to scan the scene around the vehicle, as the invisible spectrum wavelengths will not distract other road users.
Claims
1. A 3D imaging system, comprising: a light source configured to illuminate a scene with at least one light pulse, the light pulse comprising light having at least two discrete wavelengths; an optical detector including an array of detection elements; and an optical modulator comprising a variable wavelength filter inserted in the optical path between the scene and the optical detector, It is characterized in that the optical modulator is configured to differentially amplitude-modulate two wavelengths in the reflected part of the light pulse according to time changes, so that in a first operating mode the modulator differentially modulates the amplitudes of the two wavelengths, so that the first wavelength decays gradually over time and the second wavelength does not decay or decays gradually during the time, and in a second operating mode the amplitudes of the two wavelengths are modulated, so that the second wavelength decays gradually over time and the first wavelength does not decay or decays gradually.
2. The 3D imaging system of claim 1, wherein: The light source is arranged to illuminate the scene with at least two groups of light pulses, each group of pulses comprising at least one pulse comprising light having at least two discrete wavelengths; the optical detector comprising a single array of detection elements, wherein each detection element in the detector array is responsive over a wavelength range that includes the two wavelengths, the detector array producing a corresponding color image for each of the two groups of pulses; and The optical modulator is arranged to differentially modulate the two wavelengths in the reflected portion of each pulse of the group of optical pulses in a temporal manner.
3. The 3D imaging system of claim 1 , further comprising a focusing optical device that focuses a portion of the light pulses that have been reflected or scattered from an object in the scene onto an optical detector, the focusing optical device comprising at least one focusing lens located in the optical path upstream of the modulator, the focusing lens forming an image of the scene in a focal plane, and wherein the detector is located in the focal plane.
4. The 3D imaging system of claim 1, wherein the light source emits broadband white light, the broadband white light comprising the two discrete wavelengths.
5. The 3D imaging system of claim 2, wherein the spectrum in each pulse is the same.
6. A 3D imaging system as described in claim 2, wherein the detector array captures a first image of the scene corresponding to a first set of pulses, whereby for each pulse, the optical modulator operates only in the first mode, and wherein the detector captures a second image of the scene corresponding to a second set of pulses, whereby for each pulse, the modulator operates only in the second mode.
7. A 3D imaging system as claimed in claim 6, wherein the first image is captured at a different time than the second image to form a pair of images, and further comprising a signal processing circuit that determines the range of an object in the scene by comparing color values measured by each detector element in the detector array in the first image and color values measured by each corresponding detector element in the array to form the second image.
8. The 3D imaging system of claim 7, wherein the comparison constructs a 3D image of the scene.
9. The 3D imaging system of claim 1, wherein: the detector array comprising a first sub-array of detector elements, wherein each detector element is responsive to a first wavelength but not to a second wavelength, the detector array further comprising a second sub-array of detector elements, wherein each detector element is responsive to the second wavelength but not to the first wavelength, each sub-array being arranged, in use, to output a monochromatic image; and The beam splitter receives light reflected or scattered from the scene and splits it into two beams, the first beam forming an image of the scene on the first subarray and the second beam forming an image of the scene on the second subarray.
10. The 3D imaging system of claim 9, wherein the first sub-array and the second sub-array comprise color detector arrays, each color detector array being covered with a filter corresponding to a respective one of the two wavelengths.
11. The 3D imaging system of claim 9, further comprising a focusing lens located upstream of the modulator, the focusing lens forming an image of the scene onto a focal plane, and wherein the detector is located in the focal plane.
12. A method for 3D imaging of an object in a scene, comprising: illuminating the scene with at least one light pulse; differentially amplitude modulating at least two wavelengths in a portion of the pulses reflected or scattered by an object in a scene, and focusing the modulated light onto at least one detector to form an image of the scene; and processing at least one image to determine the extent of objects in the scene, The step of differentially modulating is characterized in that it includes differentially modulating the amplitudes of the two wavelengths in a first mode, so that the first wavelength decays gradually over time and the second wavelength does not decay or decays gradually during the time, and individually modulating the amplitudes of the two wavelengths in a second mode, so that the second wavelength decays gradually over time and the first wavelength does not decay or decays gradually.
Citation Information
Patent Citations
Systems and methods for high-resolution three-dimensional imaging
CN102292980A
