WO2021017051A1 - 一种图像采集芯片、物体成像识别设备及物体成像识别方法 - Google Patents

一种图像采集芯片、物体成像识别设备及物体成像识别方法 Download PDF

Info

Publication number
WO2021017051A1
WO2021017051A1 PCT/CN2019/101772 CN2019101772W WO2021017051A1 WO 2021017051 A1 WO2021017051 A1 WO 2021017051A1 CN 2019101772 W CN2019101772 W CN 2019101772W WO 2021017051 A1 WO2021017051 A1 WO 2021017051A1
Authority
WO
WIPO (PCT)
Prior art keywords
modulation
light
layer
image
sensing
Prior art date
Application number
PCT/CN2019/101772
Other languages
English (en)
French (fr)
Inventor
崔开宇
蔡旭升
朱鸿博
黄翊东
张巍
冯雪
刘仿
Original Assignee
清华大学
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 清华大学 filed Critical 清华大学
Priority to JP2020514179A priority Critical patent/JP7232534B2/ja
Priority to KR1020207005095A priority patent/KR102358804B1/ko
Priority to US16/978,137 priority patent/US11489596B2/en
Priority to EP19909643.9A priority patent/EP4007253A4/en
Publication of WO2021017051A1 publication Critical patent/WO2021017051A1/zh
Priority to US17/950,568 priority patent/US11881896B2/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L27/00—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate
    • H01L27/14—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components sensitive to infrared radiation, light, electromagnetic radiation of shorter wavelength or corpuscular radiation and specially adapted either for the conversion of the energy of such radiation into electrical energy or for the control of electrical energy by such radiation
    • H01L27/144—Devices controlled by radiation
    • H01L27/146—Imager structures
    • H01L27/14643—Photodiode arrays; MOS imagers
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
    • H04B10/50—Transmitters
    • H04B10/516—Details of coding or modulation
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
    • G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
    • G01J3/02—Details
    • G01J3/0256—Compact construction
    • G01J3/0259—Monolithic
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
    • G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
    • G01J3/12—Generating the spectrum; Monochromators
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
    • G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
    • G01J3/28—Investigating the spectrum
    • G01J3/2803—Investigating the spectrum using photoelectric array detector
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
    • G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
    • G01J3/28—Investigating the spectrum
    • G01J3/30—Measuring the intensity of spectral lines directly on the spectrum itself
    • G01J3/36—Investigating two or more bands of a spectrum by separate detectors
    • G—PHYSICS
    • G02—OPTICS
    • G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
    • G02B5/00—Optical elements other than lenses
    • G02B5/20—Filters
    • G02B5/201—Filters in the form of arrays
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L27/00—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate
    • H01L27/14—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components sensitive to infrared radiation, light, electromagnetic radiation of shorter wavelength or corpuscular radiation and specially adapted either for the conversion of the energy of such radiation into electrical energy or for the control of electrical energy by such radiation
    • H01L27/144—Devices controlled by radiation
    • H01L27/146—Imager structures
    • H01L27/14601—Structural or functional details thereof
    • H01L27/14603—Special geometry or disposition of pixel-elements, address-lines or gate-electrodes
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L27/00—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate
    • H01L27/14—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components sensitive to infrared radiation, light, electromagnetic radiation of shorter wavelength or corpuscular radiation and specially adapted either for the conversion of the energy of such radiation into electrical energy or for the control of electrical energy by such radiation
    • H01L27/144—Devices controlled by radiation
    • H01L27/146—Imager structures
    • H01L27/14601—Structural or functional details thereof
    • H01L27/14609—Pixel-elements with integrated switching, control, storage or amplification elements
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L27/00—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate
    • H01L27/14—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components sensitive to infrared radiation, light, electromagnetic radiation of shorter wavelength or corpuscular radiation and specially adapted either for the conversion of the energy of such radiation into electrical energy or for the control of electrical energy by such radiation
    • H01L27/144—Devices controlled by radiation
    • H01L27/146—Imager structures
    • H01L27/14601—Structural or functional details thereof
    • H01L27/14625—Optical elements or arrangements associated with the device
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
    • H04B10/50—Transmitters
    • H04B10/516—Details of coding or modulation
    • H04B10/54—Intensity modulation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/50—Constructional details
    • H04N23/54—Mounting of pick-up tubes, electronic image sensors, deviation or focusing coils
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/50—Constructional details
    • H04N23/55—Optical parts specially adapted for electronic image sensors; Mounting thereof

Definitions

  • the present disclosure relates to the technical field of imaging and object recognition equipment, and in particular to an image acquisition chip, an object imaging recognition device and an object imaging recognition method.
  • Imaging spectroscopy is a technology that organically combines spectral detection and imaging. It can image an object under different spectra, and at the same time obtain the geometric shape information and spectral characteristics of the detected object. In the ultraviolet, visible and near-infrared electromagnetic waves In the mid-infrared region, it is a technology to acquire many very narrow and continuous spectral image data. After more than 30 years of development, imaging spectroscopy technology has become an important means of earth observation and deep space exploration. It has been widely used in agriculture, animal husbandry and forestry production, mineral resource exploration, cultural relic detection, ocean remote sensing, environmental monitoring, disaster prevention and mitigation, Military reconnaissance and other fields.
  • Imaging spectrometer There are three imaging methods for imaging spectrometers. Among them, the optical scanning type has moving parts and is heavy and bulky; the push broom imaging type optical system is complex; the staring imaging type has a spatial resolution and the number of spectral channels. limit. None of these three can meet the needs of miniaturization and lightweight.
  • the embodiments of the present disclosure provide an image acquisition chip, an object imaging recognition device, and an object imaging recognition method to solve the defect that the imaging spectrometer in the existing image acquisition device cannot meet the requirements of miniaturization and light weight.
  • an image acquisition chip which includes a light modulation layer, an image sensing layer, and at least two sets of pixel confirmation modules.
  • the light modulation layer is located on the image sensing layer.
  • Each of the pixel confirmation modules includes a modulation unit and a sensing unit, each of the modulation unit and each of the sensing units are respectively arranged up and down on the light modulation layer and the image sensing layer; wherein, each of the modulation units At least one modulation subunit is respectively provided in each of the modulation subunits, and each modulation subunit is respectively provided with a plurality of modulation holes passing through the light modulation layer, and each modulation hole in the same modulation subunit is arranged Form a two-dimensional graphic structure with a specific arrangement rule.
  • the specific arrangement rule of the two-dimensional graphic structure includes:
  • All the modulation holes in the same two-dimensional graphic structure have the same specific cross-sectional shape at the same time, and each of the modulation holes is arranged in an array according to the gradual order of structural parameter size; and/or
  • Each of the modulation holes in the same two-dimensional pattern structure has a specific cross-sectional shape, and each of the modulation holes is arranged in combination according to the specific cross-sectional shape.
  • the arrangement sequence is arranged row by row or column by column according to a preset period sequence.
  • no modulation holes are provided in the modulation subunits at the same position in each modulation unit.
  • the bottom of the modulation hole penetrates the light modulation layer or does not penetrate the light modulation layer.
  • it further includes a signal processing circuit layer, which is connected under the image sensing layer and electrically connects the sensing units.
  • the sensing unit includes at least one sensing subunit, each of the sensing subunits is arranged in a matrix, each sensing subunit is provided with at least one image sensor, and all the sensing subunits They are electrically connected through the signal processing circuit layer.
  • a light-transmitting medium layer is further included, and the light-transmitting medium layer is located between the light modulating micro/nano structure and the image sensing layer.
  • the present disclosure also provides an object imaging recognition device, including:
  • the light source is used to emit a spectrum to the object to be imaged, so that after the spectrum passes through the object to be imaged, it is incident on the image acquisition chip as incident light;
  • the image acquisition chip as described above is arranged on the same side of the object to be imaged at the same time as the light source, and the image acquisition chip is used to use each group of pixel confirmation modules to light modulate the incident light to obtain at least two Each of the modulated spectra is detected, and the light intensity of each modulated spectrum is respectively detected, so as to determine each pixel point.
  • the present disclosure also provides a method for identifying an object to be imaged, which is proposed based on the above-mentioned object imaging and identifying device; the method for identifying an object to be imaged includes:
  • a light source to emit a spectrum to the object to be imaged, so that the spectrum passes through the object to be imaged, and is incident on the image acquisition chip as incident light;
  • Each group of pixel confirmation modules of the image acquisition chip is used to optically modulate the incident light to obtain several modulated spectra, and the light intensity of each modulated spectra is detected by sensing and respectively determining Each group of pixel data;
  • All the pixel data are integrated to form an output image.
  • the image acquisition chip described in the present disclosure includes a light modulation layer, an image sensing layer, and at least two sets of pixel confirmation modules.
  • the light modulation layer is located on the image sensing layer.
  • Each set of pixel confirmation modules includes a modulation unit and a sensing unit.
  • Each modulation unit and each sensing unit are respectively arranged on the light modulation layer and the image sensing layer up and down, and each group of pixel confirmation modules are used to detect the light intensity of the spectrum respectively to determine the pixel data of each pixel. , And then integrate all the pixel data to form the final output image.
  • the image acquisition chip can replace the complex and precise spectroscopic elements and excessive image sensors in the existing object imaging and recognition equipment.
  • the modulation unit and the sensing unit are used to respectively modulate the spectrum and sense the light intensity, thereby achieving accurate image reconstruction.
  • the structure process enables the image acquisition chip to perform light intensity sensing without the need for gratings, prisms, mirrors or other similar spatial light splitting elements, thereby greatly reducing the volume of object imaging recognition equipment and improving the light intensity sensing
  • the precision makes the object imaging recognition equipment have the advantages of high measurement accuracy, good portability, real-time online detection, simple operation, stable performance, and low manufacturing cost.
  • the application of this chip on small platforms such as small satellites and drones has broad prospects.
  • Each modulation unit in the image acquisition chip is provided with at least one modulation subunit, and each modulation subunit is provided with several modulation holes passing through the light modulation layer, and each modulation subunit in the same modulation subunit The holes are arranged into a two-dimensional graphic structure with a specific arrangement rule.
  • the chip is based on the modulation effect of the modulation unit array in optoelectronics on the light of different wavelengths, and each modulation unit is correspondingly nested with multiple modulation sub-units, so that the image information of the object to be imaged of multiple spectra can be collected at the same time. Improve the spectral recognition rate and reduce the false recognition rate.
  • the chip can be used for iris recognition.
  • the recognition rate is improved, and it can be used for anti-counterfeiting in living body detection, increasing the difficulty of cracking and forgery, and reducing contact lenses and cosmetic lenses. And interference caused by different lighting conditions.
  • This chip overcomes the problem of expensive and incapable of miniaturization of existing object imaging and recognition equipment.
  • FIG. 1 is an imaging principle diagram of an object imaging recognition device according to an embodiment of the disclosure
  • FIG. 2 is a diagram of the working principle of imaging the iris by the object imaging recognition device according to the embodiment of the disclosure
  • FIG. 3 is a schematic diagram of the structure of the image acquisition chip according to the first embodiment of the disclosure.
  • FIG. 4 is a cross-sectional view of the image acquisition chip according to the first embodiment of the disclosure.
  • FIG. 5 is a schematic diagram of a light modulation layer according to Embodiment 1 of the disclosure.
  • FIG. 6 is a schematic diagram of the structure of the image sensing layer in the first embodiment of the disclosure.
  • FIG. 7 is a schematic diagram of the structure of the light modulation layer in the second embodiment of the disclosure.
  • FIG. 8 is a schematic diagram of the structure of the light modulation layer in the third embodiment of the disclosure.
  • FIG. 9 is a schematic diagram of the structure of the light modulation layer in the fourth embodiment of the disclosure.
  • FIG. 10 is a cross-sectional view of the image acquisition chip according to the fifth embodiment of the disclosure.
  • FIG. 11 is a cross-sectional view of the image acquisition chip according to the sixth embodiment of the disclosure.
  • FIG. 12 is a schematic diagram of the structure of the image acquisition chip according to the seventh embodiment of the disclosure.
  • FIG. 13 is a schematic diagram of the structure of the light modulation layer in the seventh embodiment of the disclosure.
  • FIG. 14 and 15 are schematic diagrams of the process of the preparation method of the modulation hole processing of the image acquisition chip according to the first to seventh embodiments of the disclosure, respectively.
  • 100 light source
  • 200 object to be imaged
  • 300 image acquisition chip
  • Substrate 1. Light modulation layer; 2. Image sensing layer; 3. Signal processing circuit layer; 4. Light-transmitting medium layer; 5. Modulation unit; 6. Modulation hole; 7. Sensing unit; 8. Gap; 9. Sensing subunit; 11, first modulation unit; 12, second modulation unit; 13, third modulation unit; 14, fourth modulation unit; 15, fifth modulation unit; 16, sixth modulation unit; 110.
  • the fourth subunit 1.
  • Embodiments of the present disclosure provide an image acquisition chip, which can replace the complex and precise spectroscopic elements and excessive image sensors in the existing object imaging and recognition equipment, and use the modulation unit and the sensing unit to separately perform multiple spectra
  • the light intensity is modulated and sensed, thereby realizing a precise image reconstruction process, so that the image acquisition chip can perform light intensity sensing work without the need for gratings, prisms, mirrors or other similar spatial light splitting elements, thereby greatly reducing
  • the volume of the object imaging recognition device also improves the precision of light intensity sensing, so that the object imaging recognition device has the advantages of high measurement accuracy, good portability, real-time online detection, simple operation, stable performance, and low manufacturing cost.
  • the chip 300 includes a light modulation layer 1, an image sensing layer 2, and at least two sets of pixel confirmation modules.
  • the light modulation layer 1 is located above the image sensing layer 2.
  • the light modulation layer 1 is used to receive and modulate the incident light
  • the image sensing layer 2 is used to sense the intensity of the modulated spectrum, so as to target different
  • the light intensity of the wavelength spectrum determines the corresponding image pixel data.
  • Each group of pixel confirmation modules includes a modulation unit 5 and a sensing unit 7.
  • each group are respectively arranged on the light modulation layer 1 and the image sensing layer 2 up and down, so that each The modulation unit 5 and each sensing unit 7 can correspondingly determine at least one set of image pixel data. Integrate all the image pixel data to reconstruct the original output image.
  • each modulation unit 5 is provided with at least one modulation subunit, and each modulation subunit is respectively provided with several modulation holes 6 passing through the light modulation layer 1.
  • Each of the modulation subunits The modulation holes 6 are arranged in a two-dimensional pattern structure with a specific arrangement rule.
  • the image capture chip 300 of this embodiment can replace the precise optical components in the existing object imaging and recognition equipment to achieve precise modulation of incident light and pixel reconstruction; and the image capture chip 300 can be used to flexibly realize different wavelengths Modulation of light, including but not limited to light scattering, absorption, projection, reflection, interference, surface plasmon and resonance, etc., to improve the difference in spectral response between different regions, thereby improving the performance of the image acquisition chip 300 Analysis accuracy; in addition, the chip 300 uses one or more two-dimensional pattern structures on each modulation unit 5 to achieve modulation of light of different wavelengths, and the difference in the two-dimensional pattern structure can also improve the spectral response between different regions Differences, thereby improving the analysis accuracy of the image acquisition chip 300.
  • Modulation of light including but not limited to light scattering, absorption, projection, reflection, interference, surface plasmon and resonance, etc.
  • the chip 300 of this embodiment also includes a signal processing circuit layer 3, which is connected under the image sensing layer 2 and electrically connects the sensing units 7 so as to be able to confirm what the module senses according to each group of pixels.
  • the light intensity measurement calculates the differential response. Since the sensor in each sensing unit 7 can form a pixel according to the light intensity sensed, the intensity distribution of each wavelength on a pixel can be obtained through an algorithm.
  • the differential response refers to the difference between the response spectrum signal obtained after modulation by the light modulation layer and the original spectrum signal, or the difference between the response spectrum signals obtained after modulation by each modulation unit 5, or It is the difference between the response spectrum signals obtained after modulation of each modulation subunit.
  • the original spectrum refers to the spectrum of incident light incident on the light modulation layer 1.
  • each sensing unit 7 includes at least one sensing subunit 9, each sensing subunit 9 is arranged in a matrix, each sensing subunit 9 is provided with at least one image sensor, and all sensing subunits 9 pass through
  • the signal processing circuit layer 3 is electrically connected in order to distinguish the pixel points formed on the modulation unit 5 and the sensing unit 7 of different groups in more detail, so as to use the modulation subunits in the same position of each group and the corresponding sensing subunits 9 to generate multiple
  • Each pixel constitutes an image containing multiple spectrum information.
  • an object imaging recognition device is proposed based on various embodiments of the present disclosure.
  • the device includes a light source 100 and an image acquisition chip.
  • the light source 100 is used to emit a spectrum to the object 200 to be imaged, so that after the spectrum passes through the object 200 to be imaged, it is incident on the image acquisition chip 300 as incident light.
  • the image capture chip 300 and the light source 100 are simultaneously arranged on the same side of the object 200 to be imaged.
  • the image acquisition chip 300 is configured to use each group of pixel confirmation modules to light modulate the incident light to obtain at least two modulated spectra, and to detect the light intensity of each modulated spectra to determine each pixel. Point in order to finally integrate all the pixels to form an image.
  • the light source 100 and the image capture chip 300 are simultaneously arranged on the underside of the object 200 to be imaged.
  • the spectrum of visible-near infrared light generated by the light source 100 can penetrate the object to be measured.
  • the entire spectrum forms incident light that enters the image acquisition chip 300.
  • FIG. 2 shows an imaging principle diagram of the object imaging recognition device described in this disclosure applied to imaging the iris 210. It is understandable that the image acquisition chip 300 and the object imaging recognition device described in the present disclosure can perform imaging recognition on the iris 210, and can also perform imaging recognition on any other object 200 to be imaged. The volume of each modulation unit 5 of the collection chip 300 and the corresponding wavelength of the incident light can be adjusted.
  • the image acquisition chip 300 and the object imaging recognition device of the present disclosure will be described in detail through several embodiments.
  • the chip 300 described in each of the following embodiments is applicable to the above-mentioned object imaging recognition device.
  • the light modulation layer 1 includes a plurality of modulation units 5. All the modulation holes 6 in the modulation unit 5 penetrate the light modulation layer 1.
  • the two-dimensional image structure composed of multiple modulation holes 6 in each modulation unit 5 has the same specific cross-sectional shape.
  • the first embodiment uses six modulation units 5 composed of an array of oval modulation holes 6 as shown in FIG. Take an example. All the modulation holes 6 in each modulation unit 5 are arranged in an array according to the same regular structural parameter size gradient order to form a two-dimensional graphic structure. In this two-dimensional graphic structure, all modulation holes 6 are arranged in an array, and all modulation holes 6 are arranged row by row according to the length of the major axis, the length of the minor axis and the rotation angle from small to large.
  • the reconstruction process is implemented through a data processing module, which includes spectral data preprocessing and data prediction models.
  • the spectral data preprocessing refers to the preprocessing of the noise in the differential response data obtained above
  • the processing methods used in the spectral data preprocessing include but are not limited to Fourier transform, differentiation, and wavelet transform.
  • the data prediction model includes the prediction of the graphic parameters of the object to be imaged 200 obtained from spectral data information, and the algorithms used include but are not limited to least squares, principal component analysis, and artificial neural networks.
  • each modulation unit 5 is arranged up and down corresponding to a sensing unit 7, so that the image sensor in the sensing unit 7 forms a pixel point of the spectrum received by the modulation unit 5, and the algorithm can Get the intensity distribution on a pixel.
  • the integration of multiple pixels corresponding to different groups of pixel confirmation modules can form an image containing multiple spectrum information.
  • each modulation subunit has different modulation effects on light of different wavelengths.
  • the modulation effects include but are not limited to scattering, absorption, Transmission, reflection, interference, excimer, etc.
  • the final effect of modulation is to determine that the transmission spectra of different wavelengths of light passing through the different modulation subunits of each modulation unit 5 are different, and the same input spectrum will pass through the same The transmission spectra of different regions in the two-dimensional image structure are different.
  • each sensing unit 7 there is a sensing unit 7 under each modulation unit 5, and each sensing unit 7 is provided with multiple sensors. As shown in FIG. 6, each sensor in the same sensing unit 7 corresponds to With different areas in the same modulation unit 5, each sensor and its corresponding area constitute a pixel point. Therefore, each group of pixel confirmation modules can form more than one pixel point. Through the algorithm, each wavelength on a pixel point can be obtained. The intensity distribution. The integration of multiple pixels corresponding to the same position in different sets of pixel confirmation modules can form an image containing multiple spectrum information.
  • the above-mentioned specific cross-sectional shape of the modulation hole 6 includes a circle, an ellipse, a cross, a regular polygon, a star or a rectangle, etc., and can also be any combination of the above shapes.
  • the above-mentioned structural parameters of the modulation hole 6 include inner diameter, major axis length, minor axis length, rotation angle, number of angles, or side length.
  • the light source 100 applicable to the object imaging recognition device described in the first embodiment is a light source in the visible to near-infrared wavelength range, and the wavelength range of the light source 100 is 400 nm to 1100 nm.
  • the thickness of the light modulation layer 1 is 60 nm to 1200 nm, and the light modulation layer 1 and the image sensing layer 2 are directly connected or connected through a light-transmitting medium layer 4.
  • the image sensing layer 2 and the signal processing circuit layer 3 are electrically connected. Among them, as shown in Fig.
  • the two-dimensional pattern structure is a matrix structure as a whole, and the area of the matrix structure ranges from 200 ⁇ m 2 to 40,000 ⁇ m 2.
  • a silicon-based material is selected as the material of the light modulation layer 1 and the image sensing layer 2 at the same time, so as to have good compatibility in the manufacturing process.
  • the light modulation layer 1 can be directly generated on the image sensing layer 2, or the prepared light modulation layer 1 can be transferred to the image sensing layer 2 first.
  • the direct generation method of the light modulation layer 1 specifically includes: directly depositing on the image sensing layer 2 to generate the light modulation layer 1 arranged according to the structure shown in FIG. 5; or installing a silicon base on the image sensing layer 2 first.
  • the substrate made of material is then subjected to micro-nano processing and opening on the substrate according to the structure shown in FIG. 5 to obtain the light modulation layer 1.
  • the above-mentioned direct deposition growth process is: the first step is to deposit a silicon plate on the image sensing layer 2 by sputtering, chemical vapor deposition, or other methods.
  • the second step is to use photolithography, electron beam exposure and other graphic transfer methods to draw the required two-dimensional graphic structure on it. The structure is shown in Figure 6.
  • the two-dimensional pattern structure is specifically as follows: only the minor axis and the rotation angle of the elliptical modulation hole 6 are gradually adjusted, the major axis of the ellipse is selected from a fixed value of 200 nm to 1000 nm, for example, 500 nm; the length of the minor axis changes in the range of 120 nm to 500 nm , The rotation angle of the ellipse varies from 0° to 90°, and the arrangement period of the ellipse is a fixed value from 200 nm to 1000 nm, such as 500 nm.
  • the overall area of the pattern of the modulation unit 5 ranges from 200 ⁇ m 2 to 40,000 ⁇ m 2 in a rectangular array structure.
  • the silicon plate is etched by reactive ion etching, inductively coupled plasma etching, ion beam etching and other methods to obtain the desired light modulation layer 1.
  • the entire light modulation layer 1 and the image sensing layer 2 are electrically connected to the signal processing circuit layer 3.
  • the above-mentioned transfer preparation method of the light modulation layer 1 is specifically as follows: firstly, open holes on the substrate according to the structure shown in FIG. 5 through micro-nano processing to obtain the prepared light modulation layer 1, and then the prepared light modulation layer 1 The modulation layer 1 is transferred to the image sensing layer 2.
  • the process of the transfer method of the light modulation layer 1 is as follows: first prepare the light modulation layer 1 on a silicon wafer or SOI (referring to a silicon-insulator-silicon wafer structure) according to the above parameters, and then transfer to the image sensor through the transfer method. On layer 2, finally, the entire light modulation layer 1 and image sensing layer 2 are electrically connected to the signal processing circuit layer 3.
  • this embodiment also provides another preparation process of the image acquisition chip 300, specifically: the image sensing layer 2 is equipped with III-V group detectors, specifically GaAs/InGaAs quantum Trap detector. As shown in Fig. 14, the detector is flip-bonded on the CMOS circuit. The detector includes a GaAs substrate 1'and an InGaAs quantum well image sensing layer 2. As shown in Fig. 15, after directly thinning the substrate 1', micro-nano processing is performed on the substrate 1'to make it have a two-dimensional pattern structure, thereby forming the light modulation layer 1.
  • III-V group detectors specifically GaAs/InGaAs quantum Trap detector.
  • the detector is flip-bonded on the CMOS circuit.
  • the detector includes a GaAs substrate 1'and an InGaAs quantum well image sensing layer 2.
  • micro-nano processing is performed on the substrate 1'to make it have a two-dimensional pattern structure, thereby forming the light modulation layer 1.
  • the complete process of image acquisition and reconstruction by the object imaging recognition device of this embodiment is as follows: As shown in FIG. 2, first, a broad-spectrum light source 100 from visible light to near-infrared is irradiated on the iris 210 of the human eye, so that the iris 210 is incident on The light is absorbed and reflected on the chip 300, and the reflected light from the iris 210 to the chip 300 is the incident light of the chip 300. Then, the incident light enters the light modulation layer 1 and passes through the light modulation effect of each modulation unit 5. In this process, the modulation effect of different areas on each modulation unit 5 is different, and the transmission spectrum is also different, and each modulation unit 5 has different modulation effects.
  • each modulation unit 5 in Figure 4 Each area corresponds to each sensing subunit 9 in each sensing unit 7 in FIG. 5, and the transmission spectrum obtained by each sensing subunit 9 is different. Therefore, each modulation subunit is different from each sensing subunit 9 A group of pixel confirmation sub-modules can be formed, and each sub-module can respectively identify part of the spectrum information in a pixel. Therefore, the sub-modules of various regions can be integrated to obtain multiple spectrum information of a pixel. Further integrating each pixel point can obtain all the pixels of the image and reconstruct the iris image accordingly.
  • the structure, principle, object imaging recognition method, and chip preparation method of the image acquisition chip 300 and the object imaging recognition device of the second embodiment are basically the same as those of the first embodiment, and the similarities are not repeated here. The differences are:
  • each modulation unit 5 provided on the light modulation layer 1 has their own specific cross-sectional shapes.
  • the modulation holes 6 are arranged in free combination according to a specific cross-sectional shape (that is, random arrangement in an irregular manner).
  • a specific cross-sectional shape that is, random arrangement in an irregular manner.
  • the specific cross-sectional shapes of some modulation holes 6 are the same, and each modulation hole 6 with the same specific cross-sectional shape constitutes a plurality of modulation holes 6 groups, and the specific cross-sectional shapes of the modulation holes 6 are mutually exclusive. They are not the same, and all the modulation holes 6 are freely combined.
  • the modulation unit 5 as a whole can be regarded as modulating the spectrum of a specific wavelength, or it can be freely divided into several micro-nano modulation sub-units, so as to be able to modulate the spectrum of multiple different wavelengths. To increase the flexibility and diversity of light modulation.
  • the structure, principle, object imaging and recognition method, and chip preparation method of the image acquisition chip 300 and the object imaging recognition device of the third embodiment are basically the same as those of the second embodiment, and the similarities are not repeated here. The difference is:
  • Two or more modulation units 5 are arranged on the light modulation layer 1 of the image acquisition chip 300 of this embodiment, and each modulation unit 5 is further divided into at least two modulation sub-units. All the modulation holes 6 in each modulation subunit are arranged in combination according to a specific cross-sectional shape, and the order of arrangement is row-by-row or column-by-column arrangement according to a preset periodic sequence; the same area in each modulation unit 5 The cross-sectional shape and arrangement period of the modulation holes 6 in the modulation subunit are the same. Therefore, the modulation sub-units at different positions of each modulation unit 5 have different modulation effects on the same incident light.
  • Changing the gradual sequence of the structural parameters of the modulation hole 6 in the modulation unit 5 and/or the specific cross-sectional shape of the modulation hole 6 can change the modulation function and/or the modulation object of the current modulation unit 5 according to the modulation needs.
  • each of the three modulation units 5 is a row, a total of two rows, specifically the first modulation unit 11, The second modulation unit 12, the third modulation unit 13 and the fourth modulation unit 14, the fifth modulation unit 15, and the sixth modulation unit 16 correspondingly arranged in the second row.
  • Each modulation unit is further divided into four modulation subunits according to the same structural ratio, specifically the first subunit 110 located at the upper left corner of the unit matrix, the second subunit 111 located at the upper right corner of the unit matrix, and the second subunit 111 located at the lower left corner of the unit matrix.
  • the first subunit 112 and the first subunit 113 at the lower right corner of the unit matrix.
  • the modulation hole structure parameters and arrangement period of the modulation subunits in the same area on each modulation unit 5 are the same.
  • the modulation holes 6 in the first modulation subunit 110 and the second modulation subunit 111 are both circular, but the inner diameters of the modulation holes 6 are different, so the first modulation subunit 110 has a first modulation method for the input spectrum ,
  • the second modulation subunit 111 has a second modulation method for the input spectrum.
  • the modulation holes 6 in the third modulation subunit 112 are all elliptical, and the third modulation subunit 112 has a third modulation method for the input spectrum; the modulation holes 6 in the fourth modulation unit 113 are all triangular, the fourth The modulation holes 6 in the modulation unit 113 are arranged in a periodic row by column according to the size of the structure parameter, and the fourth modulation unit 113 has a fourth modulation method for the input spectrum.
  • the modulation subunits at the same position in different modulation units 5 have the same structure, but the modulation subunits at different positions are different from each other, so each modulation subunit has a different modulation effect on the same incident light.
  • Each modulation subunit corresponds to a sensing subunit on the image sensing layer 2 respectively.
  • the "a certain modulation method for light of different wavelengths" in this embodiment may include, but is not limited to, scattering, absorption, transmission, reflection, interference, surface plasmon, resonance, etc.
  • the first, second and third light modulation methods are different from each other.
  • the modulation effect can be changed by changing the structural parameters of the modulation holes 6 in each modulation unit 5.
  • the changes in the structural parameters include, but are not limited to, the modulation hole arrangement period of the two-dimensional pattern structure, One of the modulation hole radius and the side length, duty cycle and thickness of the modulation unit and any combination of them.
  • the duty cycle refers to the ratio of the area of the modulation hole 6 to the total area of the modulation unit 5 as a whole.
  • the light modulation layer 1 is made of a silicon nitride flat plate with a thickness of 200 nm to 500 nm.
  • a total of 1,000 to 250,000 modulation units 5 are provided on the light modulation layer 1, and the area of each modulation unit 5 ranges from 200 ⁇ m 2 to 40,000 ⁇ m 2.
  • Each modulation unit 5 selects various geometric shapes as the specific cross-sectional shape of the modulation hole 6, and each modulation unit 5 is periodically arranged in the same shape, and the duty ratio is 10% to 90%.
  • the rest of the structure is the same as Embodiment 1 or Embodiment 2.
  • Each modulation unit 5 and the sensing unit below each constitute a pixel point. Through the algorithm, the intensity distribution of each wavelength on a pixel can be obtained. By integrating the sub-modules at the same position of different units to form an image of the same modulation mode of pixels, multiple pixels form an image containing multiple spectrum information.
  • each modulation unit 5 has a modulation subunit in the same area, and there is no modulation hole 6 in the modulation subunit.
  • the first modulation sub-unit 110, the second modulation sub-unit 111, and the third modulation sub-unit 112 respectively correspond to a certain wavelength of light with a narrow-band filtering effect, while the fourth modulation sub-unit 113 does not If the modulation hole 6 is provided, the incident light directly passes through the area of the fourth modulation subunit 113.
  • a corresponding sensing sub-unit is provided below each modulation sub-unit, and the light passes through the narrowband filtering of the first modulation sub-unit 110, the second modulation sub-unit 111, and the third modulation sub-unit 112, respectively
  • the light sensor in the corresponding sensing subunit detects the light intensity, and since the light passing through the fourth modulation subunit 113 has no narrow-band filtering effect, the light intensity detected by the corresponding sensing subunit 9 can be used as a comparison item.
  • the first three groups of light intensities are separately processed with the fourth group of light intensities to obtain the light intensity of each wavelength after narrowband filtering.
  • the fourth modulation subunit 113 can also be used to locate the boundary of the object.
  • micro-integrated image acquisition chip 300 of this embodiment can use the modulation unit 5 of the first embodiment or the modulation unit 5 of the second embodiment or the modulation unit 5 of the third embodiment or the first and the first embodiment. Any combination of the modulation unit 5 described in the second and the third embodiment.
  • the fifth embodiment proposes an image capture chip 300, an object imaging recognition device and an object imaging recognition methods.
  • the similarities between the fifth embodiment and the foregoing embodiments will not be repeated, and the differences are:
  • the image capture chip 300 of the fifth embodiment further includes a light-transmitting medium layer 4, and the light-transmitting medium layer 4 is located between the light modulation layer 1 and the image sensing layer 2.
  • the thickness of the transparent medium layer 4 is 50 nm to 1 ⁇ m, and the material may be silicon dioxide.
  • the image sensing layer 2 can be covered by chemical vapor deposition, sputtering, spin coating, etc.
  • the optical medium layer 4 is then deposited and etched on the part of the optical modulation layer 1.
  • silicon dioxide can be used as the preparation substrate of the light modulation layer 1, and the light modulation layer 1 can be directly prepared by micro-nano drilling on the upper half of the substrate, and then the silicon dioxide substrate
  • the lower half of ⁇ is directly used as the light-transmitting medium layer 4, and the prepared light modulation layer 1 and the light-transmitting medium layer 4 are transferred to the image sensing layer 2 as a whole.
  • the light-transmitting medium layer 4 of this embodiment can also be configured as follows: the entire light modulation layer 1 above the image sensing layer 2 is supported by an external support structure so that it is suspended relative to the image sensing layer 2, then the light modulation layer The air part between the layer 1 and the image sensing layer 2 is the light-transmitting medium layer 4.
  • the sixth embodiment further proposes an image capture chip 300, an object imaging recognition device And object imaging recognition method.
  • the similarities between this sixth embodiment and the above-mentioned embodiments will not be repeated, and the differences are:
  • each modulation hole 6 does not penetrate the light modulation layer. It is understandable that no matter whether the modulation hole 6 penetrates the light modulation layer or not, it will not adversely affect the modulation effect of the light modulation layer 1, because the silicon-based materials or other materials selected for the light modulation layer 1 are all light-transmitting materials. When the spectrum is incident on the light modulation layer 1, it is affected by the structure of each modulation unit 5 to produce a modulation effect, but the bottom of the modulation hole 6 does not adversely affect the spectrum modulation.
  • the thickness from the bottom of the modulation hole 6 of the light modulation layer 1 to the bottom of the light modulation layer is 60 nm to 1200 nm, and the thickness of the entire light modulation layer is 120 nm to 2000 nm.
  • the seventh embodiment further proposes an image acquisition chip 300, an object imaging recognition device And object imaging recognition method.
  • the similarities between the seventh embodiment and the foregoing embodiments will not be repeated here, and the differences are:
  • the image acquisition chip 300 of the seventh embodiment six modulation units 5 are distributed on the light modulation layer of the light modulation layer 1, which are respectively a first modulation unit 11 and a second modulation unit 12. ,
  • the first modulation unit 11, the third modulation unit 13, and the fourth modulation unit 14 adopt the periodic structure of the modulation unit 5 described in the third embodiment, and the second modulation unit 12, the fifth modulation unit 15 and the sixth modulation unit 16
  • the gradual configuration of the modulation unit 5 described in the first embodiment is adopted.
  • the modulation unit 5 at the corresponding position can select any suitable structure of the modulation unit 5 in the first to fourth embodiments.
  • the light modulation layer 1 of the seventh embodiment uses the difference in the specific cross-sectional shape of the different modulation holes 6 between different units and the specific arrangement of the modulation holes 6 in the same unit to realize the use of changing the specific cross-sectional shape of the modulation holes 6,
  • the structural parameters of the modulation holes 6 and the arrangement period of the modulation holes 6 realize different modulation effects on spectra of different wavelengths.
  • the arbitrarily divided modulation units 5 have different modulation effects on the spectrum, and theoretically, an infinite number of modulated groups can be obtained.
  • Spectral samples greatly increase the amount of data used to reconstruct the original spectrum, which helps to restore the spectral pattern of the broadband spectrum.
  • the periodic structure can produce two-dimensional periodic dispersion and resonance.
  • the resonance includes but not limited to the band control of the photonic crystal and the resonance of the two-dimensional grating. principle.
  • the resonance effect can enhance the detection accuracy for a specific wavelength.
  • the above two advantages can be combined.
  • the image acquisition chip 300 of the above three embodiments can be prepared into a structure of the order of micrometers or even smaller, which is great for the miniaturization and miniaturization of the micro-integrated image acquisition chip 300.
  • the use of the chip 300 is of great significance.
  • the overall size of the chip 300 is equivalent to that of a camera module, which can be less than 1cm ⁇ 1cm ⁇ 0.4cm.
  • the chip 300 can be integrated in portable mobile devices such as mobile phones and bracelets.
  • the above-mentioned light modulation layer 1 cooperates with the image sensing layer 2 composed of different image sensors, in principle, the full-wavelength spectrum detection can be achieved, so that the image acquisition chip 300 has a better broadband detection performance.
  • the eighth embodiment further proposes a miniature spectral imaging chip 300 and a spectral imaging device And spectral imaging methods.
  • the target object 2 can be extended to any object. As shown in FIG. 1, first, a broad-spectrum light source 100 from visible light to near-infrared is irradiated on the target object 200, and then the reflected light is collected by the image acquisition chip 300. Or omit the light source 100 to emit light directly from the target object 200 to the image capture chip 300 for collection.
  • each modulation unit 5 the incident light enters the light modulation layer 1 and passes through the light modulation effect of each modulation unit 5.
  • the modulation effect of different regions on each modulation unit 5 is different, and the transmission spectrum It is also different, and there are multiple sensing units 7 on the image sensing layer 2 under each modulation unit, and as shown in Figures 4 and 5, it can be seen that under the corresponding action of multiple sets of pixel confirmation modules, the Each area in each modulation unit 5 corresponds to each sensing subunit 9 in each sensing unit 7 in FIG. 5, and the transmission spectrum obtained by each sensing subunit 9 is different, so each modulation subunit is Each sensing sub-unit 9 can form a set of pixel confirmation sub-modules.
  • Each sub-module can respectively identify part of the spectrum information in a pixel. Therefore, the sub-modules of each area can be integrated to obtain multiple spectrum information of a pixel. . Further integrating each pixel point can obtain all the pixels of the image and reconstruct the target object image accordingly. It is understandable that since the two-dimensional graphic structure on the modulation unit corresponding to each sensing subunit 9 is the same, the response of light at different spatial positions of the image after the same modulation action is obtained, and the target object image at the same frequency can be obtained.
  • the image acquisition chip 300 includes a light modulation layer 1, an image sensing layer 2, and at least two sets of pixel confirmation modules.
  • the light modulation layer 1 is located on the image sensing layer 2.
  • Each group of pixel confirmation modules includes a modulation unit 5 and a sensing unit 7.
  • Each modulation unit 5 and each sensing unit 7 are respectively arranged up and down on the light modulation layer 1 and the image sensing layer 2, and each group of pixel confirmation modules are used respectively.
  • the image acquisition chip 300 can replace the complex and precise spectroscopic elements and excessive image sensors in the existing object imaging and recognition equipment.
  • the modulation unit 5 and the sensing unit 7 are used to respectively modulate the spectrum and sense the light intensity, thereby achieving accurate
  • the image reconstruction process enables the image acquisition chip 300 to perform light intensity sensing work without the need for gratings, prisms, mirrors or other similar spatial light-splitting elements, thereby greatly reducing the volume of the object imaging recognition equipment and improving the
  • the precision of light intensity sensing makes the object imaging recognition equipment have the advantages of high measurement accuracy, good portability, real-time online detection, simple operation, stable performance, and low manufacturing cost.
  • Each modulation unit in the image acquisition chip 300 is provided with at least one modulation sub-unit, and each modulation sub-unit is respectively provided with a plurality of modulation holes 6 passing through the light modulation layer 1, and each of the modulation sub-units
  • the modulation holes 6 are arranged in a two-dimensional pattern structure with a specific arrangement rule.
  • the chip 300 is based on the modulation effect of the modulation unit 5 array in optoelectronics on light of different wavelengths, and each modulation unit 5 is correspondingly nested with multiple modulation sub-units, so that it can collect multiple spectrum object image information at the same time. It greatly improves the spectral recognition rate and reduces the misrecognition rate.
  • the biometric iris recognition can be used for anti-counterfeiting, and it can increase the difficulty of cracking and forgery. At the same time, it can also reduce the contact lens, cosmetic contact lenses and different light conditions.
  • the interference overcomes the problem that the existing object imaging and recognition equipment is expensive and cannot be miniaturized.
  • connection should be interpreted broadly. For example, they can be fixed or detachable. Connected or integrally connected; it can be mechanically or electrically connected; it can be directly connected or indirectly connected through an intermediate medium.
  • connection should be interpreted broadly. For example, they can be fixed or detachable. Connected or integrally connected; it can be mechanically or electrically connected; it can be directly connected or indirectly connected through an intermediate medium.
  • connection should be interpreted broadly. For example, they can be fixed or detachable. Connected or integrally connected; it can be mechanically or electrically connected; it can be directly connected or indirectly connected through an intermediate medium.

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • General Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Condensed Matter Physics & Semiconductors (AREA)
  • Computer Hardware Design (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Multimedia (AREA)
  • Optics & Photonics (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)
  • Studio Devices (AREA)
  • Transforming Light Signals Into Electric Signals (AREA)
  • Die Bonding (AREA)

Abstract

一种图像采集芯片（300）、物体成像识别设备及物体成像识别方法，图像采集芯片（300）的各组像素确认模块中，每个调制单元（5）和每个感应单元（7）分别上下对应的设置在光调制层（1）和图像传感层（2）上，每个调制单元（5）内分别设有至少一个调制子单元（110、111、112、113），每个调制子单元（110、111、112、113）内分别设有若干个穿于光调制层内的调制孔（6），同一调制子单元（110、111、112、113）内的各个调制孔（6）排布成一具有特定排布规律的二维图形结构。图像采集芯片（300）基于光电子学中的调制单元（5）阵列对不同波长光的调制作用，且能同时采集多个光谱的图像信息，克服了现有物体成像识别设备价格昂贵且无法小型化的问题。

Description

一种图像采集芯片、物体成像识别设备及物体成像识别方法
相关申请的交叉引用
本申请要求于2019年7月31日提交的申请号为201910700328.7，发明名称为“一种图像采集芯片、物体成像识别设备及物体成像识别方法”的中国专利申请的优先权，其通过引用方式全部并入本申请。
技术领域
本公开涉及成像及物体识别设备技术领域，尤其涉及一种图像采集芯片、物体成像识别设备及物体成像识别方法。
背景技术
成像光谱学
成像光谱学是一种将光谱探测和成像有机结合的技术，能够对某一物体进行不同光谱下的成像，同时获得被探测物体的几何形状信息和光谱特征，在电磁波的紫外、可见光、近红外和中红外区域，获取许多非常窄且光谱连续的图像数据的技术。经过三十多年的发展，成像光谱技术已经成为对地观测和深空探测的重要手段，被广泛应用于农牧林生产、矿产资源勘查、文物检测、海洋遥感、环境监测、防灾减灾、军事侦察等领域。
随着搭载平台的逐步小型化，如小卫星、无人机等小型平台，以及野外应用的续航需求，对图像采集设备小型化、轻量化的需求越来越突出。传统的图像采集设备通常为成像光谱仪，成像光谱仪有三种成像方式，其中光机扫描式具有运动部件，重量重体积大；推帚成像式光学系统复杂；凝视成像式空间分辨率及光谱通道数受限。这三者都无法满足小型化、轻量化的需求。
发明内容
(一)要解决的技术问题
本公开实施例提供了一种图像采集芯片、物体成像识别设备及物体成像识别方法，用以解决现有的图像采集设备中的成像光谱仪不能满足小型化、轻量化需求的缺陷。
(二)技术方案
为了解决上述技术问题，本公开提供了一种图像采集芯片，包括光调制层、图像传感层以及至少两组像素确认模块，所述光调制层位于图像传感层的上面，每组所述像素确认模块均包括调制单元和感应单元，每个所述调制单元和每个所述感应单元分别上下对应的设置在所述光调制层和图像传感层上；其中，每个所述调制单元内分别设有至少一个调制子单元，每个所述调制子单元内分别设有若干个穿于所述光调制层内的调制孔，同一所述调制子单元内的各个所述调制孔排布成一具有特定排布规律的二维图形结构。
在部分实施例中，所述二维图形结构的特定排布规律包括：
同一所述二维图形结构内的所有所述调制孔同时具有相同的特定截面形状，各个所述调制孔按照结构参数大小渐变顺序成阵列排布；和/或
同一所述二维图形结构内的各个所述调制孔分别具有特定截面形状，各个所述调制孔按照特定截面形状进行组合排列。
在部分实施例中，各个所述调制孔按照特定截面形状进行组合排列时，所述排列的顺序为按照预设周期顺序逐行或逐列排布。
在部分实施例中，每个所述调制单元中的同一位置的所述调制子单元内不设有调制孔。
在部分实施例中，所述调制孔的底部穿透所述光调制层或是不穿透所述光调制层。
在部分实施例中，还包括信号处理电路层，连接在所述图像感应层的下面，并将各个所述感应单元之间电连接。
在部分实施例中，所述感应单元包括至少一个感应子单元，各个所述感应子单元成矩阵排列，每个所述感应子单元内分别设有至少一个图像传感器，所有的所述感应子单元之间分别通过所述信号处理电路层电连接。
在部分实施例中，还包括透光介质层，所述透光介质层位于所述光调制微纳结构与图像传感层之间。
本公开还提供了一种物体成像识别设备，包括：
光源，用于向待成像物体发射光谱，以使所述光谱经过所述待成像物体后，作为入射光射入图像采集芯片上；以及
如上所述的图像采集芯片，与所述光源同时设置在所述待成像物体的同一侧，所述图像采集芯片用于利用各组像素确认模块分别对所述入射光进行光调制以得到至少两个调制后的光谱，并对各个所述调制后的光谱的光强分别进行感应探测，从而分别确定各个像素点。
本公开还提供了一种待成像物体识别方法，是基于如上所述的物体成像识别设备提出的；该待成像物体识别方法包括：
利用光源向待成像物体发射光谱，以使所述光谱经过所述待成像物体后，作为入射光射入图像采集芯片上；
利用所述图像采集芯片的各组像素确认模块分别对所述入射光进行光调制以得到若干个调制后的光谱，并对各个所述调制后的光谱的光强分别进行感应探测，从而分别确定各组像素数据；
将所有的所述像素数据整合，以构成输出图像。
(三)有益效果
本公开的上述技术方案具有以下有益效果：
1、本公开所述的图像采集芯片包括光调制层、图像传感层以及至少两组像素确认模块，光调制层位于图像传感层的上面，每组像素确认模块均包括调制单元和感应单元，每个调制单元和每个感应单元分别上下对应的设置在光调制层和图像传感层上，利用各组像素确认模块分别对光谱进行光强感应探测，从而分别确定各像素点的像素数据，进而整合所有像素数据以构成最终的输出图像。该图像采集芯片能取代现有的物体成像识别设备中的复杂精密的分光元件和过多的图像传感器，利用调制单元和感应单元分别对光谱进行调制并感应光强，从而实现了精确的图像重构过程，使得图像采集芯片能在不需要光栅、棱镜、反射镜或其他类似空间分光元件的情况下进行光强感应工作，进而大大缩小了物体成像识别设备的体积，同时提高了光强感应的精密性，使得物体成像识别设备具有测量精度高、便携性好、可实时在线检测、操作简单、性能稳定、制造成本低等优点。该芯片在小型平台如小卫星、无人机等上的应用有广阔的前景。
2、该图像采集芯片中每个调制单元内分别设有至少一个调制子单元，每个调制子单元内分别设有若干个穿于光调制层内的调制孔，同一调制子单元内的各个调制孔排布成一具有特定排布规律的二维图形结构。该芯片 基于光电子学中的调制单元阵列对不同波长光的调制作用，且每个调制单元中对应嵌套有多个调制子单元，从而能同时采集多个光谱的待成像物体图像信息，极大提高了光谱识别率、降低误识率。该芯片可用于虹膜识别，通过同时获得多个波长下的虹膜图像信息，提高识别率，并且能有利于活体检测用于防伪，增大破解、伪造难度，同时还可以减小隐形眼镜、美瞳和不同光照条件下造成的干扰。该芯片克服了现有物体成像识别设备价格昂贵且无法小型化的问题。
附图说明
为了更清楚地说明本公开实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图是本公开的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本公开实施例的物体成像识别设备的成像原理图；
图2为本公开实施例的物体成像识别设备对虹膜进行成像的工作原理图；
图3为本公开实施例一的图像采集芯片的结构示意图；
图4为本公开实施例一的图像采集芯片的剖视图；
图5为本公开实施例一的光调制层示意图；
图6为本公开实施例一的图像感应层的结构示意图；
图7为本公开实施例二的光调制层的结构示意图；
图8为本公开实施例三的光调制层的结构示意图；
图9为本公开实施例四的光调制层的结构示意图；
图10为本公开实施例五的图像采集芯片的剖视图；
图11为本公开实施例六的图像采集芯片的剖视图；
图12为本公开实施例七的图像采集芯片的结构示意图；
图13为本公开实施例七的光调制层的结构示意图；
图14和图15分别为本公开实施例一至实施例七的图像采集芯片调制孔加工制备方法的过程示意图。
其中，100、光源；200、待成像物体；300、图像采集芯片；
1’、衬底；1、光调制层；2、图像传感层；3、信号处理电路层；4、 透光介质层；5、调制单元；6、调制孔；7、感应单元；8、间隙；9、感应子单元；11、第一调制单元；12、第二调制单元；13、第三调制单元；14、第四调制单元；15、第五调制单元；16、第六调制单元；110、第一子单元；111、第二子单元；112、第三子单元；113、第四子单元。
具体实施方式
下面结合附图和实施例对本公开的实施方式作进一步详细描述。以下实施例用于说明本公开，但不能用来限制本公开的范围。除非另有说明，否则本公开中所提及的芯片均为图像采集芯片的简称。
本公开各实施例提供了一种图像采集芯片，该芯片能取代现有的物体成像识别设备中的复杂精密的分光元件和过多的图像传感器，利用调制单元和感应单元分别对多个光谱进行同时调制并感应光强，从而实现了精确的图像重构过程，使得图像采集芯片能在不需要光栅、棱镜、反射镜或其他类似空间分光元件的情况下进行光强感应工作，进而大大缩小了物体成像识别设备的体积，同时提高了光强感应的精密性，使得物体成像识别设备具有测量精度高、便携性好、可实时在线检测、操作简单、性能稳定、制造成本低等优点。
具体的，如图1～图13所示，该芯片300包括光调制层1、图像传感层2以及至少两组像素确认模块。光调制层1位于图像传感层2的上面，光调制层1用于接收入射光并对该入射光进行调制，而图像传感层2用于感应调制后的光谱的光强，从而针对不同波长的光谱的光强确定对应的图像像素点数据。每组的像素确认模块均包括调制单元5和感应单元7，对应的，每组的调制单元5和感应单元7分别上下对应设置在光调制层1和图像传感层2上，从而使得每个调制单元5和每个感应单元7都能对应的确定至少一组图像像素点数据。将所有的图像像素点数据整合，即可重构出原输出图像。
本实施例中，每个调制单元5内分别设有至少一个调制子单元，每个调制子单元内分别设有若干个穿于光调制层1内的调制孔6，同一调制子单元内的各个调制孔6排布成一具有特定排布规律的二维图形结构。本实施例的图像采集芯片300能够取代现有的物体成像识别设备中的精密光学部件，以实现对入射光的精密调制和像素重构；并且利用该图像采集芯片 300可以灵活地实现对不同波长光的调制作用，该调制作用包括但不限于光的散射、吸收、投射、反射、干涉、表面等离子激元以及谐振等作用，提高不同区域间光谱响应的差异性，从而提高图像采集芯片300的分析精度；此外，该芯片300利用每个调制单元5上的一个或多个二维图形结构实现对不同波长的光的调制作用，利用二维图形结构的区别还可以提高不同区域间光谱响应的差异性，从而提高图像采集芯片300的分析精度。
本实施例的芯片300还包括信号处理电路层3，信号处理电路层3连接在图像感应层2的下面，并将各个感应单元7之间电连接，从而能够根据各组像素确认模块感应到的光强测算差分响应，由于每个感应单元7内的传感器都能根据感应到的光强而构成一个像素点，则通过算法可得到一个像素点上各个波长的强度分布。该差分响应是指经过光调制层调制后得到的响应光谱的信号与原光谱的信号之间求差值，或是对各个调制单元5调制后得到的响应光谱的信号之间求差值，或是对各个调制子单元调制后得到的响应光谱的信号之间求差值。原光谱是指射入光调制层1的入射光光谱。
进一步的，每个感应单元7包括至少一个感应子单元9，各个感应子单元9成矩阵排列，每个感应子单元9内分别设有至少一个图像传感器，所有的感应子单元9之间分别通过信号处理电路层3电连接，以便于更细致的区分不同组的调制单元5和感应单元7上构成的像素点，从而利用各组相同位置的调制子单元和对应的感应子单元9生成的多个像素点构成一幅包含多个频谱信息的图像。
如图1所示，基于本公开各实施例提出了一种物体成像识别设备，该设备包括光源100和图像采集芯片。光源100用于向待成像物体200发射光谱，以使光谱经过待成像物体200后，作为入射光射入图像采集芯片300上。图像采集芯片300与光源100同时设置在待成像物体200的同一侧。该图像采集芯片300用于利用各组像素确认模块分别对入射光进行光调制以得到至少两个调制后的光谱，并对各个调制后的光谱的光强分别进行感应探测，从而分别确定各个像素点，以便最终将所有像素点进行整合，进而构成图像。
为了便于使光源100的光谱穿过待成像物体200从而使形成的反射光 作为芯片300的入射光，优选将光源100与图像采集芯片300同时设置在待成像物体200的同一侧。以图2所示为例，将光源100与图像采集芯片300同时设置在待成像物体200的下侧，利用光的反射原理，光源100产生的可见-近红外光的光谱能穿入待测体200内，并在反射作用下全部光谱形成射入图像采集芯片300中的入射光。该结构设置能够扩大检测空间，提高物体成像和识别的使用便利性。
图2所示即为本公开所述的物体成像识别设备应用在对虹膜210进行成像的成像原理图。可理解的是，本公开所述的图像采集芯片300和物体成像识别设备既可对虹膜210进行成像识别，也可对其他任一待成像物体200进行成像识别，只需根据成像需要，对图像采集芯片300的各个调制单元5的体积以及入射光的对应波长等参数进行调整即可。
以下以虹膜210作为成像对象为例，具体通过若干个实施例对本公开的图像采集芯片300及物体成像识别设备进行详细说明。下述各个实施例所述的芯片300均适用于上述的物体成像识别设备。
实施例一
如图3和图4所示，本实施例一提供的图像采集芯片300中，光调制层1包括多个调制单元5。该调制单元5内的所有调制孔6均贯穿光调制层1。每个调制单元5内由多个调制孔6组成的二维图像结构都具备相同的特定截面形状，本实施例一以图2所示的六个由椭圆形调制孔6阵列组成的调制单元5为例进行说明。每个调制单元5内的所有调制孔6都是按照同样规律的结构参数大小渐变顺序成阵列排布形成二维图形结构。该二维图形结构中，所有调制孔6成阵列排布，并且所有调制孔6按照长轴长度、短轴长度和旋转角度由小到大逐行逐列排布。
可理解的是，如图5所示，由于本实施例的所有调制孔6都是按照同一排布规律进行排列的，即按照长轴长度、短轴长度和旋转角度的结构参数由小到大逐行逐列的渐变排布，故而该光调制层1上的所有调制孔6既可以视为一整体调制单元5，也可以进一步将其任意分割成若干个调制单元5，任意划分出的调制单元5对于光谱都有不同的调制作用，理论上可获得无穷多组调制后的光谱样本，从而急剧增大了用以重构原光谱的数据量，有助于对于宽带光谱的谱型的恢复。则根据每个调制单元5内的调制 孔6结构参数特性确定该调制单元5对不同波长的光的调制作用的效果即可。该重构过程通过数据处理模块实施，数据处理模块包括光谱数据预处理以及数据预测模型。其中，光谱数据预处理是指对上述求得的差分响应数据中存在的噪声进行预处理，该光谱数据预处理所采用的处理方法包括但不限于傅里叶变换、微分和小波变换等。数据预测模型中包括由光谱数据信息得到对待成像物体200的图形参数的预测，其使用的算法包括但不限于最小二乘法、主成分分析以及人工神经网络。
可理解的是，本实施例将每个调制单元5与一个感应单元7相对应的上下设置，以便利用感应单元7内的图像传感器将该调制单元5接收的光谱构成一个像素点，通过算法可得到一个像素点上的强度分布。而不同组的像素确认模块对应构成的多个像素点整合可构成一幅包含多个频谱信息的图像。
进一步的，由于每个调制单元5内划分的不同区域(调制子单元)的结构参数不同，则每个调制子单元对于不同波长的光的调制作用不同，调制作用包括但不限于散射、吸收、透射、反射、干涉、激元等作用，调制作用的最终效果是确定不同波长的光透过每个调制单元5的不同调制子单元区域时的透射谱不同，则针对同一输入光谱在经过同一二维图像结构中不同区域的透射谱不同。
本实施例中，每个调制单元5的下方分别对应有一个感应单元7，每个感应单元7内分别设有多个传感器，如图6所示，同一感应单元7内的每个传感器分别对应着同一调制单元5内的不同区域，则每个传感器及其对应的区域分别构成一个像素点，故而每组像素确认模块内分别能够构成超过一个像素点，通过算法可得到一个像素点上各个波长的强度分布。而不同组的像素确认模块中相同位置所对应构成的多个像素点整合可构成一幅包含多个频谱信息的图像。
可理解的是，上述的调制孔6的特定截面形状包括圆形、椭圆形、十字形、正多边形、星形或矩形等，也可以为上述各形状的任意组合。则对应的，上述的调制孔6的结构参数包括内径、长轴长度、短轴长度、旋转角度、角数或边长等。
本实施例一所述的物体成像识别设备中适用的光源100为可见光到近 红外波段的光源，光源100的波长范围为400nm～1100nm。光调制层1的厚度为60nm～1200nm，光调制层1与图像感应层2之间直接连接或者通过透光介质层4连接。图像感应层2与信号处理电路层3之间为电连接。其中，如图5所示，光探测层上共设有六个调制单元5，所有的调制单元5成阵列排布，每个调制单元5中的所有调制孔6均为椭圆形，所有椭圆形调制孔6的短轴长度分别逐行逐列增大，并且以图5中水平向为横轴，竖向为纵轴，则每个调制单元5中的所有椭圆形调制孔6逐行逐列的自纵轴向横轴旋转，其旋转角度逐渐增大。每个调制单元5中的所有调制孔6各自组成了相同的二维图形结构，该二维图形结构整体为一矩阵结构，该矩阵结构的面积范围为200μm2～40000μm2。
本实施例的图像采集芯片300在制造时，选用硅基材料同时作为光调制层1和图像感应层2的材料，以便在制备工艺的加工上具有很好的兼容性。在制备光调制层1时，可直接在图像感应层2上生成光调制层1，也可以先将已制备好的光调制层1转移至图像感应层2上。
具体的，光调制层1的直接生成方式具体包括：直接在图像感应层2上沉积生成按照图5所示的结构排布的光调制层1；或是先在图像感应层2上装有硅基材料制成的衬底，然后在衬底上按照图5所示的结构进行微纳加工开孔，以得到光调制层1。
上述的直接沉积生长的过程为：第一步、在图像感应层2上通过溅射、化学气相沉积等方法沉积硅平板。第二步、用光刻、电子束曝光等图形转移方法在上面绘制出所需的二维图形结构，结构如图6所示。该二维图形结构具体为：仅对椭圆形调制孔6的短轴和旋转角度进行渐变调整，椭圆长轴选取200nm～1000nm中的定值，例如500nm；短轴长度在120nm～500nm范围内变化，椭圆的旋转角度在0°～90°范围内变化，椭圆的排列周期为200nm～1000nm中的定值，例如500nm。调制单元5的图形整体面积范围为200μm2～40000μm2的矩形阵列结构。第三步、通过反应离子刻蚀、感应耦合等离子体刻蚀以及离子束刻蚀等方法对硅平板进行刻蚀即可得到所需光调制层1。最后将光调制层1和图像感应层2整体通过电连接到信号处理电路层3上即可。
上述的光调制层1的转移制备方式具体为：先在衬底上按照图5所示 的结构通过微纳加工开孔，以得到制备好的光调制层1，然后将该已制备好的光调制层1转移到图像感应层2上。具体的，光调制层1的转移方法的过程为：先根据以上参数在硅片或SOI(指硅-绝缘体-硅片结构)上制备得到光调制层1，然后通过转移的方法转移到图像感应层2上，最后将光调制层1和图像感应层2整体通过电连接到信号处理电路层3上即可。
如图14和图15所示，本实施例还给出了另一种图像采集芯片300的制备过程，具体为：图像感应层2内装有III-V族探测器，具体为GaAs/InGaAs的量子阱探测器。如图14所示，将探测器倒扣键合在CMOS电路上，探测器包括GaAs衬底1’和InGaAs量子阱图像感应层2。如图15所示，直接将衬底1’进行减薄后，再在衬底1’上进行微纳加工，使之具有二维图形结构，从而形成光调制层1即可。该制备过程与上述的微纳加工开孔的区别仅在于将由探测器组成的图像感应层2的上表面直接作为威纳加工的衬底1’，从而保证了加工制备好的光调制层1与图像感应层2之间的紧密连接，避免出现缝隙影响光的调制作用效果。
本实施例的物体成像识别设备对于图像的采集重构的完整流程为：如图2所示，首先，令可见光到近红外的宽谱光源100照射人眼虹膜210，以使虹膜210对射入光吸收并反射到芯片300上，虹膜210射向芯片300上的反射光即为芯片300的入射光。然后，该入射光射入光调制层1并通过各个调制单元5的光调制作用，在此过程中，各个调制单元5上的不同区域的调制作用不同，则透射谱也不同，且每个调制单元下方分别对应有图像感应层2上的多个感应单元7，并且如图4和图5所示可知，在多组像素确认模块的对应作用下，图4中的每个调制单元5中的各个区域分别与图5中的每个感应单元7中的各个感应子单元9相对应，则每个感应子单元9获得的透射谱不同，故而每个调制子单元与每个感应子单元9分别能构成一组像素确认子模块，每个子模块能分别识别一个像素点内的一部分频谱信息，故而将各个区域的子模块整合即可得到一个像素点的多个频谱信息。进一步整合各个像素点即可获得该图像所有像素点并据此重构得到虹膜图像。可理解的是，由于各个感应子单元9对应的调制单元上的二维图形结构相同，则获得图像不同空间位置的光经过相同调制作用之后的响应，可获得同一频率下的虹膜图像。
实施例二
本实施例二的图像采集芯片300及物体成像识别设备的结构、原理、物体成像识别方法和芯片制备方法均与实施例一基本相同，相同之处不再赘述，不同之处在于：
如图7所示，本实施例的图像采集芯片300中，光调制层1上设有的各个调制单元5中，各个二维图形结构内的所有调制孔6分别具有各自的特定截面形状，各个调制孔6按照特定截面形状进行自由组合排列(即无规律的任意排列)。具体的，在该二维图形结构内，部分调制孔6的特定截面形状相同，具有相同特定截面形状的各个调制孔6构成了多个调制孔6组，各个调制孔6组的特定截面形状互不相同，且所有的调制孔6均自由组合。
可理解的是，该调制单元5整体可视为针对一种特定波长的光谱进行调制，也可以将其自由分割成若干个微纳调制子单元，从而能针对多种不同波长的光谱进行调制，以增加光调制的灵活性和多样性。
实施例三
本实施例三的图像采集芯片300及物体成像识别设备的结构、原理、物体成像识别方法和芯片制备方法均与实施例二基本相同，相同之处不再赘述。不同之处在于：
本实施例的图像采集芯片300的光调制层1上排列有两个或两个以上的调制单元5，每个调制单元5中进一步分割为至少两个调制子单元。每个调制子单元内的所有调制孔6分别按照特定截面形状进行组合排列，且其排列的顺序为按照预设的周期顺序逐行或逐列排布；每个调制单元5内的同样区域的调制子单元内的调制孔6的截面形状和排布周期相同。从而使得每个调制单元5的不同位置的调制子单元对同一入射光都具有不同的调制作用。根据调制需要改变调制单元5内的调制孔6结构参数的渐变顺序和/或调制孔6的特定截面形状，即可改变当前调制单元5的调制作用和/或调制对象。
具体如图8所示，光调制层1的光调制层上分布有六个调制单元5，每三个调制单元5为一行，共计两行，具体分别为第一行的第一调制单元11、第二调制单元12、第三调制单元13以及对应排列在第二行的第四调 制单元14、第五调制单元15、第六调制单元16。每个调制单元分别按照同样结构比例进一步划分为四个调制子单元，具体分别为位于单元矩阵左上角的第一子单元110、位于单元矩阵右上角的第二子单元111、位于单元矩阵左下角的第一子单元112和位于单元矩阵右下角的第一子单元113。
其中，每个调制单元5上相同区域的调制子单元的调制孔结构参数和排列周期均相同。具体的，第一调制子单元110和第二调制子单元111内的调制孔6均为圆形，但调制孔6的内径不同，则第一调制子单元110对于输入光谱有第一种调制方式，第二调制子单元111对于输入光谱有第二种调制方式。第三调制子单元112内的调制孔6均为椭圆形，该第三调制子单元112对于输入光谱有第三种调制方式；第四调制单元113内的调制孔6均为三角形，该第四调制单元113内的各个调制孔6按照结构参数大小成周期式逐行逐列排列，则该第四调制单元113对于输入光谱有第四种调制方式。则不同调制单元5中相同位置的调制子单元结构相同，但不同位置的调制子单元相互不同，故而每个调制子单元对同一入射光的调制作用不同。每个调制子单元分别对应于图像感应层2上的一个感应子单元。
可以理解的是，本实施例的“对不同波长的光有某种调制方式”可包括但不限于散射、吸收、透射、反射、干涉、表面等离子激元、谐振等作用。第一、第二和第三种光调制方式彼此区别。通过对于调制单元5内的调制孔6结构的设置，可以提高不同单元间光谱响应的差异，通过增加单元数量就可以提高对不同谱之间差异的灵敏度。
可理解的是，针对不同入射光谱测量时，可通过改变各调制单元5内的调制孔6结构参数来改变调制作用，结构参数的改变包括但不限于二维图形结构的调制孔排布周期、调制孔半径以及调制单元的边长、占空比和厚度等各参数中的一种以及它们的任意组合。其中，占空比是指调制孔6的面积和调制单元5整体总面积之比。
本实施例中，光调制层1为厚度200nm～500nm的氮化硅平板制成。光调制层1上共设有1000～250000个调制单元5，每个调制单元5的面积范围为200μm2～40000μm2。每个调制单元5内部选取各种几何形状作为调制孔6的特定截面形状，每个调制单元5内为同一形状的周期排布，其占空比为10％～90％。其余结构均与实施例1或实施例2相同。
每个调制单元5分别与其下方的感应单元构成一个像素点。通过算法可以得到一个像素点上各个波长的强度分布。通过整合不同单元的相同位置的子模块构成像素点的同一调制方式下的图像，多个像素点构成一幅包含多个频谱信息的图像。
实施例四
本实施例四的图像采集芯片300及物体成像识别设备的结构、原理、物体成像识别方法和芯片制备方法均与实施例三基本相同，相同之处不再赘述。不同之处在于：每个调制单元5上分别有一相同区域上的调制子单元内不设有任何调制孔6。以图9所示为例，第一调制子单元110、第二调制子单元111和第三调制子单元112分别对应某一个特定波长的光有窄带滤波作用，而第四调制子单元113中不设有调制孔6，则入射光直接通过该第四调制子单元113的区域。
对应的，每个调制子单元的下方分别设有对应的感应子单元，则光通过第一调制子单元110、第二调制子单元111和第三调制子单元112各自的窄带滤波后，分别由对应的感应子单元内的光传感器探测到光强，而通过第四调制子单元113的光由于没有存在窄带滤波作用，则其对应的感应子单元9探测到的光强可作为对比项。将前三组光强分别与第四组光强进行差分处理，从而可以得到每一个波长的经过窄带滤波后的光强。此外，设置第四调制子单元113还可以用于对物体的边界进行定位。
可理解的是，本实施例的微集成图像采集芯片300中可使用如实施例一的调制单元5或实施例二的调制单元5或实施例三的调制单元5或是实施例一、实施例二和实施例三所述的调制单元5的任意组合。
实施例五
基于上述任一实施例的图像采集芯片300及物体成像识别设备的结构、原理、物体成像识别方法和芯片制备方法，本实施例五提出了一种图像采集芯片300、物体成像识别设备以及物体成像识别方法。本实施例五与前述各实施例之间的相同之处不再赘述，不同之处在于：
如图10所示，本实施例五的图像采集芯片300还包括透光介质层4，该透光介质层4位于光调制层1与图像感应层2之间。具体的，该透光介质层4的厚度为50nm～1μm，材料可为二氧化硅。
本实施例的微集成图像采集芯片300中，若在制备光调制层1时采用直接沉积生长的工艺方案，可在图像感应层2上通过化学气相沉积、溅射以及旋涂等方式覆盖该透光介质层4，然后在其上方进行光调制层1部分的沉积、刻蚀即可。若采用转移的工艺方案，则可将二氧化硅作为光调制层1的制备衬底，并在衬底上半部分直接通过微纳钻孔加工制备光调制层1，然后以二氧化硅衬底的下半部分直接作为透光介质层4，将制备好的光调制层1与透光介质层4这两部分整体转移到图像感应层2上即可。
可理解的是，本实施例的透光介质层4还可以设置为：将图像感应层2上方的光调制层1整体通过外部支撑结构支撑以使之相对于图像感应层2悬空，则光调制层1与图像感应层2之间的空气部分即为透光介质层4。
实施例六
基于上述任一实施例所述的图像采集芯片300及物体成像识别设备的结构、原理、物体成像识别方法和芯片制备方法，本实施例六进一步提出了一种图像采集芯片300、物体成像识别设备以及物体成像识别方法。本实施例六与上述各实施例的相同之处不再赘述，不同之处在于：
如图11所示，本实施例六的图像采集芯片300中，各个调制孔6均不穿透光调制层。可理解的是，不论调制孔6是否穿透光调制层均不会对光调制层1的调制作用造成不利影响，这是因为光调制层1选用的硅基材料或其他材料均为透光材料，光谱入射光调制层1时，受到各个调制单元5的结构影响而发生调制作用，但调制孔6底部对于光谱调制不产生不利影响。
本实施例的图像采集芯片300中，光调制层1的调制孔6底部至光调制层底部的厚度为60nm～1200nm，整个光调制层厚度为120nm～2000nm。
实施例七
基于上述任一实施例所述的图像采集芯片300及物体成像识别设备的结构、原理、物体成像识别方法和芯片制备方法，本实施例七进一步提出了一种图像采集芯片300、物体成像识别设备以及物体成像识别方法。本实施例七与上述各实施例的相同之处不再赘述，不同之处在于：
如图12和图13所示，本实施例七的图像采集芯片300中，光调制层1的光调制层上分布有六个调制单元5，分别为第一调制单元11、第二调 制单元12、第三调制单元13、第四调制单元14、第五调制单元15和第六调制单元16，其中。第一调制单元11、第三调制单元13和第四调制单元14采用实施例三所述的调制单元5的周期结构设置，而第二调制单元12、第五调制单元15和第六调制单元16采用实施例一所述的调制单元5的渐变结构设置。
可理解的是，也可以根据光谱调制需要，在对应位置的调制单元5选择实施例一至实施例四中任一合适的调制单元5的结构设置即可。由此可见，本实施例七的光调制层1利用不同单元间的不同调制孔6特定截面形状的区别、以及同一单元内特定的调制孔6排列方式，实现利用改变调制孔6特定截面形状、调制孔6的结构参数以及调制孔6排列周期实现对不同波长的光谱进行不同的调制作用。
可理解的是，对于实施例一和实施例二的渐变式阵列调制单元5的结构，其任意划分出的调制单元5对于光谱都有不同的调制作用，理论上可获得无穷多组调制后的光谱样本，从而急剧增大了用以重构原光谱的数据量，有助于对于宽带光谱的谱型的恢复。
对于实施例三和实施例四的周期式调制单元5的结构，其周期结构可产生二维周期的色散、谐振作用，谐振作用包括但不限于光子晶体的能带控制以及二维光栅的谐振等原理。通过谐振作用可增强对于特定波长的探测精度。
如果将上述的实施例一、实施例二、实施例三和实施例四中的调制单元5同时应用在芯片300上时，能够综合上述两种优势。并且在切顶光调制层的尺寸范围时，上述三个实施例的图像采集芯片300都可以制备成微米量级甚至更小的结构，这对于微集成图像采集芯片300的小型化微型化生产和使用具有重大意义，芯片300整体尺寸与相机模组相当，可做到小于1cm×1cm×0.4cm的规格，该芯片300可集成在手机、手环等便携式移动设备上。另外，上述的光调制层1配合由不同的图像感应器构成的图像感应层2，在原则上可以实现对于全波段的光谱探测，从而使得图像采集芯片300的宽谱探测性能更加出色。
实施例八
基于上述任一实施例所述的图像采集芯片300及物体成像识别设备的 结构、原理、物体成像识别方法和芯片制备方法，本实施例八进一步提出了一种微型光谱成像芯片300、光谱成像设备以及光谱成像方法。目标物体2可拓展为任意物体。如图1所示，首先，令可见光到近红外的宽谱光源100照射到目标物体200上，然后反射光由图像采集芯片300采集。或省略光源100由目标物体200直接发光照射到图像采集芯片300采集。然后，如图3和图4，该入射光射入光调制层1并通过各个调制单元5的光调制作用，在此过程中，各个调制单元5上的不同区域的调制作用不同，则透射谱也不同，且每个调制单元下方分别对应有图像感应层2上的多个感应单元7，并且如图4和图5所示可知，在多组像素确认模块的对应作用下，图4中的每个调制单元5中的各个区域分别与图5中的每个感应单元7中的各个感应子单元9相对应，则每个感应子单元9获得的透射谱不同，故而每个调制子单元与每个感应子单元9分别能构成一组像素确认子模块，每个子模块能分别识别一个像素点内的一部分频谱信息，故而将各个区域的子模块整合即可得到一个像素点的多个频谱信息。进一步整合各个像素点即可获得该图像所有像素点并据此重构得到目标物体图像。可理解的是，由于各个感应子单元9对应的调制单元上的二维图形结构相同，则获得图像不同空间位置的光经过相同调制作用之后的响应，可获得同一频率下的目标物体图像。
综上所述，本公开各实施例所述的图像采集芯片300包括光调制层1、图像传感层2以及至少两组像素确认模块，光调制层1位于图像传感层2的上面，每组像素确认模块均包括调制单元5和感应单元7，每个调制单元5和每个感应单元7分别上下对应的设置在光调制层1和图像传感层2上，利用各组像素确认模块分别对光谱进行光强感应探测，从而分别确定各像素点的像素数据，进而整合所有像素数据以构成最终的输出图像。该图像采集芯片300能取代现有的物体成像识别设备中的复杂精密的分光元件和过多的图像传感器，利用调制单元5和感应单元7分别对光谱进行调制并感应光强，从而实现了精确的图像重构过程，使得图像采集芯片300能在不需要光栅、棱镜、反射镜或其他类似空间分光元件的情况下进行光强感应工作，进而大大缩小了物体成像识别设备的体积，同时提高了光强感应的精密性，使得物体成像识别设备具有测量精度高、便携性好、可实 时在线检测、操作简单、性能稳定、制造成本低等优点。
该图像采集芯片300中每个调制单元内分别设有至少一个调制子单元，每个调制子单元内分别设有若干个穿于光调制层1内的调制孔6，同一调制子单元内的各个调制孔6排布成一具有特定排布规律的二维图形结构。该芯片300基于光电子学中的调制单元5阵列对不同波长光的调制作用，且每个调制单元5中对应嵌套有多个调制子单元，从而能同时采集多个光谱的物体图像信息，极大提高了光谱识别率、降低误识率，用于生物虹膜识别能有利于活体检测用于防伪，增大破解、伪造难度，同时还可以减小隐形眼镜、美瞳和不同光照条件下造成的干扰，克服了现有物体成像识别设备价格昂贵且无法小型化的问题。
本公开的实施例是为了示例和描述起见而给出的，而并不是无遗漏的或者将本公开限于所公开的形式。很多修改和变化对于本领域的普通技术人员而言是显而易见的。选择和描述实施例是为了更好说明本公开的原理和实际应用，并且使本领域的普通技术人员能够理解本公开从而设计适于特定用途的带有各种修改的各种实施例。
在本公开的描述中，除非另有说明，“多个”和“若干个”的含义都是指两个或两个以上；除非另有说明，“缺口状”的含义为除截面平齐外的形状。术语“上”、“下”、“左”、“右”、“内”、“外”、“前端”、“后端”、“头部”、“尾部”等指示的方位或位置关系为基于附图所示的方位或位置关系，仅是为了便于描述本公开和简化描述，而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作，因此不能理解为对本公开的限制。此外，术语“第一”、“第二”、“第三”等仅用于描述目的，而不能理解为指示或暗示相对重要性。
在本公开的描述中，需要说明的是，除非另有明确的规定和限定，术语“安装”、“相连”、“连接”应做广义理解，例如，可以是固定连接，也可以是可拆卸连接，或一体地连接；可以是机械连接，也可以是电连接；可以是直接相连，也可以通过中间媒介间接相连。对于本领域的普通技术人员而言，可以具体情况理解上述术语在本公开中的具体含义。

Claims (10)

  1. 一种图像采集芯片，其特征在于，包括光调制层、图像传感层以及至少两组像素确认模块，所述光调制层位于图像传感层的上面，每组所述像素确认模块均包括调制单元和感应单元，每个所述调制单元和每个所述感应单元分别上下对应的设置在所述光调制层和图像传感层上；
    其中，每个所述调制单元内分别设有至少一个调制子单元，每个所述调制子单元内分别设有若干个穿于所述光调制层内的调制孔，同一所述调制子单元内的各个所述调制孔排布成一具有特定排布规律的二维图形结构。
  2. 根据权利要求1所述的图像采集芯片，其特征在于，所述二维图形结构的特定排布规律包括：
    同一所述二维图形结构内的所有所述调制孔同时具有相同的特定截面形状，各个所述调制孔按照结构参数大小渐变顺序成阵列排布；和/或
    同一所述二维图形结构内的各个所述调制孔分别具有特定截面形状，各个所述调制孔按照特定截面形状进行组合排列。
  3. 根据权利要求2所述的图像采集芯片，其特征在于，各个所述调制孔按照特定截面形状进行组合排列时，所述排列的顺序为按照预设周期顺序逐行或逐列排布。
  4. 根据权利要求1所述的图像采集芯片，其特征在于，每个所述调制单元中的同一位置的所述调制子单元内不设有调制孔。
  5. 根据权利要求1所述的图像采集芯片，其特征在于，所述调制孔的底部穿透所述光调制层或是不穿透所述光调制层。
  6. 根据权利要求1-5任一项所述的图像采集芯片，其特征在于，还包括信号处理电路层，连接在所述图像感应层的下面，并将各个所述感应单元之间电连接。
  7. 根据权利要求6所述的图像采集芯片，其特征在于，所述感应单元包括至少一个感应子单元，各个所述感应子单元成矩阵排列，每个所述感应子单元内分别设有至少一个图像传感器，所有的所述感应子单元之间分别通过所述信号处理电路层电连接。
  8. 根据权利要求1-5任一项所述的图像采集芯片，其特征在于，还包 括透光介质层，所述透光介质层位于所述光调制微纳结构与图像传感层之间。
  9. 一种物体成像识别设备，其特征在于，包括：
    光源，用于向待成像物体发射光谱，以使所述光谱经过所述待成像物体后，作为入射光射入图像采集芯片上；以及
    如权利要求1-8任一项所述的图像采集芯片，与所述光源同时设置在所述待成像物体的同一侧，所述图像采集芯片用于利用各组像素确认模块分别对所述入射光进行光调制以得到至少两个调制后的光谱，并对各个所述调制后的光谱的光强分别进行感应探测，从而分别确定各个像素点。
  10. 一种待成像物体识别方法，其特征在于，是基于如权利要求9所述的物体成像识别设备提出的；该待成像物体识别方法包括：
    利用光源向待成像物体发射光谱，以使所述光谱经过所述待成像物体后，作为入射光射入图像采集芯片上；
    利用所述图像采集芯片的各组像素确认模块分别对所述入射光进行光调制以得到若干个调制后的光谱，并对各个所述调制后的光谱的光强分别进行感应探测，从而分别确定各组像素数据；
    将所有的所述像素数据整合，以构成输出图像。
PCT/CN2019/101772 2019-07-31 2019-08-21 一种图像采集芯片、物体成像识别设备及物体成像识别方法 WO2021017051A1 (zh)

Priority Applications (5)

Application Number Priority Date Filing Date Title
JP2020514179A JP7232534B2 (ja) 2019-07-31 2019-08-21 画像取得チップ、物体イメージング認識設備、及び物体イメージング認識方法
KR1020207005095A KR102358804B1 (ko) 2019-07-31 2019-08-21 이미지 수집 칩, 물체 이미징 인식 장치 및 물체 이미징 인식 방법
US16/978,137 US11489596B2 (en) 2019-07-31 2019-08-21 Image collection chip, object imaging recognition device and object imaging recognition method
EP19909643.9A EP4007253A4 (en) 2019-07-31 2019-08-21 IMAGE COLLECTION CHIP, OBJECT IMAGING AND RECOGNITION DEVICE, AND OBJECT IMAGING AND RECOGNITION METHOD
US17/950,568 US11881896B2 (en) 2019-07-31 2022-09-22 Image collection chip, object imaging recognition device and object imaging recognition method

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910700328.7 2019-07-31
CN201910700328.7A CN110381243B (zh) 2019-07-31 2019-07-31 一种图像采集芯片、物体成像识别设备及物体成像识别方法

Related Child Applications (2)

Application Number Title Priority Date Filing Date
US16/978,137 A-371-Of-International US11489596B2 (en) 2019-07-31 2019-08-21 Image collection chip, object imaging recognition device and object imaging recognition method
US17/950,568 Continuation US11881896B2 (en) 2019-07-31 2022-09-22 Image collection chip, object imaging recognition device and object imaging recognition method

Publications (1)

Publication Number Publication Date
WO2021017051A1 true WO2021017051A1 (zh) 2021-02-04

Family

ID=68257341

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/101772 WO2021017051A1 (zh) 2019-07-31 2019-08-21 一种图像采集芯片、物体成像识别设备及物体成像识别方法

Country Status (7)

Country Link
US (2) US11489596B2 (zh)
EP (1) EP4007253A4 (zh)
JP (1) JP7232534B2 (zh)
KR (1) KR102358804B1 (zh)
CN (1) CN110381243B (zh)
TW (2) TWI790639B (zh)
WO (1) WO2021017051A1 (zh)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110381243B (zh) * 2019-07-31 2024-03-12 清华大学 一种图像采集芯片、物体成像识别设备及物体成像识别方法
CN111811651A (zh) * 2020-07-23 2020-10-23 清华大学 光谱芯片、光谱仪及光谱芯片制备方法
CN112018141B (zh) * 2020-08-14 2023-11-28 清华大学 基于不同形状单元的微型光谱芯片
CN112018140A (zh) * 2020-08-14 2020-12-01 清华大学 基于随机形状单元的微型光谱芯片
CN114519872A (zh) * 2020-11-18 2022-05-20 北京与光科技有限公司 指纹活体识别装置以及指纹模组
WO2022105506A1 (zh) * 2020-11-18 2022-05-27 北京与光科技有限公司 光谱成像芯片和设备及信息处理方法、以及指纹活体识别装置和指纹模组

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106052865A (zh) * 2016-05-09 2016-10-26 中国科学院微电子研究所 光学分光器及其制备方法、图像传感器以及图像成像系统
CN106847849A (zh) * 2016-12-30 2017-06-13 中国科学院西安光学精密机械研究所 一种基于超表面窄带滤光的多光谱芯片及其制备方法
CN206584063U (zh) * 2016-11-25 2017-10-24 湖南宏动光电有限公司 一种基于表面等离激元的光谱成像微滤光片
CN107894625A (zh) * 2017-09-29 2018-04-10 扬中市恒海电子科技有限公司 一种集成式红外带通滤波器及其制造方法和光谱仪
CN108007568A (zh) * 2017-12-19 2018-05-08 湖南宏动光电有限公司 一种光谱成像型微滤光片及其制备方法
CN109029726A (zh) * 2018-05-25 2018-12-18 西北工业大学 一种窗口集成式光谱/偏振成像系统
US20190101577A1 (en) * 2016-12-05 2019-04-04 Eagle Technology LLC Method and system for radio frequency (rf) spectral imager on an integrated circuit
CN109764964A (zh) * 2019-02-26 2019-05-17 中国科学院西安光学精密机械研究所 一种推扫式偏振光谱成像微系统、成像方法及制备方法

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7426040B2 (en) * 2004-08-19 2008-09-16 University Of Pittsburgh Chip-scale optical spectrum analyzers with enhanced resolution
US7145124B2 (en) * 2004-09-15 2006-12-05 Raytheon Company Multispectral imaging chip using photonic crystals
JP5760811B2 (ja) * 2011-07-28 2015-08-12 ソニー株式会社 固体撮像素子および撮像システム
CN102564586B (zh) * 2012-01-09 2013-08-07 南京邮电大学 衍射孔阵列结构微型光谱仪及其高分辨率光谱复原方法
KR101974576B1 (ko) * 2012-04-12 2019-05-02 삼성전자주식회사 대면적을 갖는 투과형 광 이미지 변조기 및 그 제조 방법과 투과형 광 이미지 변조기를 포함하는 광학장치
JP6105728B2 (ja) * 2013-06-14 2017-03-29 シャープ株式会社 固体撮像装置
JP2015232599A (ja) * 2014-06-09 2015-12-24 ソニー株式会社 光学フィルタ、固体撮像装置、および電子機器
GB2580298B (en) * 2018-11-12 2021-08-11 Dualitas Ltd A spatial light modulator for holographic projection
CN110381243B (zh) * 2019-07-31 2024-03-12 清华大学 一种图像采集芯片、物体成像识别设备及物体成像识别方法
CN210112118U (zh) * 2019-07-31 2020-02-21 清华大学 一种图像采集芯片及物体成像识别设备

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106052865A (zh) * 2016-05-09 2016-10-26 中国科学院微电子研究所 光学分光器及其制备方法、图像传感器以及图像成像系统
CN206584063U (zh) * 2016-11-25 2017-10-24 湖南宏动光电有限公司 一种基于表面等离激元的光谱成像微滤光片
US20190101577A1 (en) * 2016-12-05 2019-04-04 Eagle Technology LLC Method and system for radio frequency (rf) spectral imager on an integrated circuit
CN106847849A (zh) * 2016-12-30 2017-06-13 中国科学院西安光学精密机械研究所 一种基于超表面窄带滤光的多光谱芯片及其制备方法
CN107894625A (zh) * 2017-09-29 2018-04-10 扬中市恒海电子科技有限公司 一种集成式红外带通滤波器及其制造方法和光谱仪
CN108007568A (zh) * 2017-12-19 2018-05-08 湖南宏动光电有限公司 一种光谱成像型微滤光片及其制备方法
CN109029726A (zh) * 2018-05-25 2018-12-18 西北工业大学 一种窗口集成式光谱/偏振成像系统
CN109764964A (zh) * 2019-02-26 2019-05-17 中国科学院西安光学精密机械研究所 一种推扫式偏振光谱成像微系统、成像方法及制备方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4007253A4 *

Also Published As

Publication number Publication date
KR20210015738A (ko) 2021-02-10
US20230025329A1 (en) 2023-01-26
JP7232534B2 (ja) 2023-03-03
US11881896B2 (en) 2024-01-23
CN110381243B (zh) 2024-03-12
TWI790639B (zh) 2023-01-21
JP2022503280A (ja) 2022-01-12
TW202107068A (zh) 2021-02-16
TWI731505B (zh) 2021-06-21
KR102358804B1 (ko) 2022-02-08
EP4007253A1 (en) 2022-06-01
TW202136751A (zh) 2021-10-01
US20220085891A1 (en) 2022-03-17
CN110381243A (zh) 2019-10-25
US11489596B2 (en) 2022-11-01
EP4007253A4 (en) 2023-04-19

Similar Documents

Publication Publication Date Title
TWI731505B (zh) 一種圖像採集晶片、物體成像識別設備及物體成像識別方法
CN111490060A (zh) 光谱成像芯片及光谱识别设备
CN111505820A (zh) 单片集成的图像传感芯片及光谱识别设备
TWI741440B (zh) 一種光學調變微奈米結構、微集成光譜儀及光譜調變方法
TW201907140A (zh) 具有亮度增強之光學成像傳輸器
KR102332190B1 (ko) 비침습 혈당 측정기 및 혈당 측정 방법
TWI798940B (zh) 光譜成像晶片及訊息處理方法
CN210376122U (zh) 一种光调制微纳结构及微集成光谱仪
WO2022166189A1 (zh) 光人工神经网络智能芯片及制备方法
CN211627935U (zh) 单片集成的图像传感芯片及光谱识别设备
CN211828773U (zh) 光谱成像芯片及光谱识别设备
CN210112118U (zh) 一种图像采集芯片及物体成像识别设备
CN211122509U (zh) 光谱仪结构及电子设备
Sarwar et al. Miniaturizing a chip-scale spectrometer using local strain engineering and total-variation regularized reconstruction
CN111854949A (zh) 弱光光谱检测芯片及弱光光谱检测方法
CN207717226U (zh) 基于周期性干涉膜系fp腔扫描的高光谱成像装置
CN118158507A (zh) 图像采集芯片、物体成像识别设备及物体成像识别方法
CN112730267A (zh) 光谱仪结构及电子设备
CN212363424U (zh) 弱光光谱检测芯片
Lewis Evolution of EO/IR technology and systems

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2020514179

Country of ref document: JP

Kind code of ref document: A

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19909643

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2019909643

Country of ref document: EP

Effective date: 20220228