Optical detection method and light detection structure
The optical detection method using a multi-spectral sensor and micro control unit addresses the limitations of current sensory systems by enabling accurate and reliable contactless measurement of physical properties, such as moisture and material composition, in a compact and energy-efficient manner, suitable for diverse applications including robotics and agriculture.
Patent Information
- Application Number
- PCT/EP2024/061136
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-03
- Filing Date
- 2024-04-23
- Publication Date
- 2024-11-07
AI Technical Summary
Current sensory systems in devices face constraints in building space, power consumption, computing capabilities, and desired accuracy, limiting their effectiveness in applications such as robots and IoT devices, which require more accurate and reliable detection methods, especially for physical properties like moisture and material composition.
An optical detection method utilizing a sensor with multiple spectral channels for both visible and infrared light ranges, combined with a micro control unit for data processing, allows for contactless measurement of physical properties by analyzing reflected or transmitted light, enabling improved accuracy and reliability while reducing hardware size and power consumption.
This approach enables reliable detection of moisture and material composition with reduced hardware size and power consumption, enhancing the accuracy and reliability of sensory systems in devices, particularly in applications like robotics and agriculture, by combining spectral color sensing with moisture sensing in a single platform.
Smart Images

Figure EP2024061136_07112024_PF_FP_ABST
Abstract
Description
[0001] Optical detection method and light detection structure
[0002] DESCRIPTION
[0003] TECHNICAL FIELD
[0004] The invention relates to an optical detection method . It also relates to a corresponding light detection structure .
[0005] BACKGROUND
[0006] Since the year 2000 the number and types of sensors in everyday devices have increased drastically . With the digitali zation of many appliances , machines , and the introduction of whole new device classes as smartphones , robots for housekeeping, digital aids in agriculture , automated manufacturing, and the internet of things , more circuit-level detection systems and sensory capabilities are required and employed .
[0007] Sensory capabilities are supporting many applications and are usually constrained by building space , power consumption, computing capabilities , hardware requirements , and desired accuracy . In many cases sensory capabilities are the essence of enabling use cases in areas that are accessible for anybody, such as robots for housekeeping and other consumer-centered products .
[0008] During the last two decades of increasing demand and development ef forts those constraints were constantly reduced in a step-by- step ef fort . Today we have moisture , temperature and gyroscopic sensors in every smartphone , and their si ze is since reduced by several orders of magnitude while capabilities and cross functional uses were developed and are more desirable than ever .
[0009] With current capabilities and the development of neural networks and algorithms for machine learning or arti ficial intelligence , we reached a point were synergies in these fields with sensory equipment need to be developed to further progress and reduce the constraints on detection methods and sensors . With the above-mentioned exposure of everyday li fe to internet of things devices and larger hardware such as sel f-driving cars , lawn mowing robots , restaurant service robots and cleaning robots , there is a serious demand for more accurate and utmost reliable detection . Thus , any form of improvement is welcome and helps securing the usage of such devices . Therefore , obj ect of much development work is the improvement on detection equipment and detection methods in general and especially in terms of accuracy, fail-safe usage while still keeping or reducing building space and all other constraints mentioned above .
[0010] SUMMARY
[0011] An obj ective of the present invention is to provide an improved detection method and a detection structure for such an improved detection method .
[0012] More speci fically, a sensing platform or detection structure and a corresponding method shall be given which with simple means allow for reliably deriving a number of physical properties , in particular moisture and / or material composition, from an obj ect by contactless measurement .
[0013] This obj ective is solved by a detection method based on reflective light detection and a detection structure in accordance with the invention, as characteri zed by the features of the independent claims . Advantageous embodiments and developments are the subj ect-matter of the dependent claims and the subsequent description .
[0014] With respect to the detection method, the essence of the invention is as follows .
[0015] Provided is an optical detection method based on the detection and analysis of light reflected from an illuminated obj ect and / or of light passing through a transmissive ( transparent ) obj ect , wherein a ) an optical sensor detects spectral data from a first spectral range and a second spectral range , the second spectral range being di f ferent from the first spectral range , b ) the sensor sends the detected spectral data to a micro control unit which receives and processes the spectral data from the optical sensor, and wherein the micro control unit and / or a connected processing unit derives a number of physical properties of the obj ect from said spectral data .
[0016] In general , a sensor comprises a chip with sensor arrays for multiple optical channels and equipment to make the sensor work and pass the data generated by the sensor on to a micro control or processing unit . Such a commonly used sensor is for example known as AS7341 ( an eleven-channel spectral color sensor by AMS ) and usually comprises all electronics needed to detect and, i f necessary, emit light , while having a small build space , for example 3 mm x 4 mm x 4 mm .
[0017] In case of the present invention, said sensor preferably comprises any number or even all of the following features : Such as up to 12 light filters or spectral channels , in particular 3 or more spectral channels / light filters in the visible light spectrum for color applications , 6 to 9 spectral channels / light filters in the (near- ) infrared spectrum for spectroscopy applications , increased sensitivity and reliability based on redundancy, I2C interfaces with, for instance , 1MHz clock frequency and interrupt ( INT ) , on-chip LED driver to drive an internal or external LED ( LDR) , a low package si ze with 3 . 1 mm x 2 mm x 1 mm, a typical supply voltage of 1 . 8V (VDD) , the Pin2Pin standard configuration for adaptability, and / or an internal or external light source in optical arrangement towards the measurement obj ect , thus enabling implementation flexibility for the overall system ( reflective / transmissive setups ) . Hence , the hardware components are configured to send light towards a measurement obj ect and to detect reflected ( or transmissive ) light with said equipment for the spectral analysis by the various channels .
[0018] A micro control unit (MCU often also j ust MC ) is a small computer, preferably on a single integrated circuit . It contains one or more CPUs (processor cores ) along with memory and programmable input / output peripherals . Program memory and storage memory as well as random-access memory (RAM) are also comprised . Micro control units are designed for embedded applications . This is in contrast to the microprocessors used in PCs or other general-purpose applications consisting of various discrete chips . A micro control unit may be part of a system on a chip ( SoC ) . An SoC may connect the external micro control unit chips as the motherboard components , but an SoC usually integrates the advanced peripherals like graphics processors and interfaces for connections like Wi-Fi adapters or mobile network adapters and interface controllers as its internal micro control unit circuits .
[0019] A processing unit typically is a SoC with more compounds and / or a PC or large computing center . It can be physically connected or be connected to the micro control unit through remote connections like Wi-Fi , Bluetooth, mobile networks , or a combination of those and may by arranged in a cloud architecture or larger local host or device .
[0020] In one embodiment of the invention the detection structure and method can have an energy saving loop in which the connectivity of the MCU (power sourcing, e . g . , Wi-Fi adaptor, Bluetooth, ethernet , mobile network receiver and / or others ) is only activated i f there is a detection event , a preset activation time or schedule and pre-set or machine learning adj usted time schedule . The MCU is also physically configured to carry out these tasks . In another embodiment , with a broadband light source integrated in the detection structure , the powering of the broadband light source can be trigged by the same logic as above and lead to energy consumption reduction . In this case the MCU preferably is also configured to enable the energy management and turn the light source on and of f .
[0021] In a preferred embodiment , the micro control unit , or a connected processing unit , cross-references and / or correlates the spectral data from the first spectral range and the second spectral range to each other and thereby improves a detection result ( e . g . , in accuracy, reliability, or capability) .
[0022] In one embodiment the optic sensor detects and analyzes a fist spectral range with wavelengths from 300 nm to 690 nm and a second spectral range with wavelengths from 690 nm to 3000 nm.
[0023] In general, said two spectral ranges may have an overlap, but preferably they are essentially distinct from each other, with no or just a small overlap.
[0024] This allows for color (visible light) detection and infrared detection, in particular near infrared (NIR) , or short wavelength infrared (SWIR) detection.
[0025] More specifically, this configuration advantageously allows for the combination of the distinct and vast detection and compound specific databases for infrared light to be combined with visible light (color) databases to improve detection limits and accuracy. The storage and identification data of materials can be configured in minute details with one spectral profile being assigned to a single or specific material or substance (e.g., copper, iron, silver, gold, lithium, graphene, nano graphene rods, balls or cones, cotton, linen, viscose, wool, angora, alpaca, polyester, lycra, moleskin, fur, feathers, specific plant species, specific molds) or it can be assigned to a general category (e.g., metals, natural fabrics, synthetic fabrics, plant categories, weed categories, sprouts and / or others) .
[0026] Depending on the desired detection result or usage, the wavelength ranges offer options for specific materials to be detected and different databases of reference results to be readily used. This also enables use of a different energy profile for the method or a hardware configured to use it.
[0027] In another embodiment of the invention the micro control unit or the connected processing unit improves the processing result from the second spectral range by processing the first spectrum and referencing the results. In another embodiment of the invention the sensor detects and analyzes the first spectral range by three or more optical filters or light filters and the second spectral range by three or more light filters and sends the detection data to the micro control unit . Preferably, there are even 3-5 , 6- 8 , or 6- 9 spectral channels or light filters for detection .
[0028] The multiple channels or filters help improving the accuracy and can be better matched or improved with the visible light detection in comparison to fewer filters . Also , certain filters are better at true / false testing in correlation with speci fic visible light filters and their detected signals .
[0029] In another embodiment of the invention, the spectral detection data of the second spectral range are processed and used, in particular referenced to a database , to detect a moisture level , with the help of the micro control unit or the connected processing unit ( s ) .
[0030] This will enable moisture detection and tracking of the development of moisture i f more than one point in time is measured or detected and correlated to another .
[0031] In another embodiment of the invention the referenced data of the second spectral range are cross-referenced by or correlated to the data of the first spectral range which preferably also is processed and / or referenced to a database to detect a moisture level . This enables a more reliable detection and fail-safe detection in the borderline environments of detection limits .
[0032] In another embodiment of the invention a machine learning and / or arti ficial intelligence algorithm is used by the micro control unit and / or by the connected processing unit to cross-reference and improve the detection .
[0033] This enables improvement over many state-of-the-art detection devices without investing in additional equipment or structures and can even improve existing structures and methods . The connectivity of the MCU to a local host or cloud architecture by the before-mentioned structure improves capabilities for algorithm-based machine learning or provides Al improvements over current methods .
[0034] In another embodiment of the invention the detection method is applied or executed over time , and a change of detected spectral properties over time is measured and / or stored by the micro control unit and / or the connected processing unit .
[0035] This allows for time resolution of the detection and has many use cases such as over-time moisture control , mapping and documenting degradation of substances or matter, growing of plants , cooking of meals , completion of a chemical reaction of process , etc .
[0036] In another embodiment of the invention, the micro control unit and / or the connected processing unit correlates a change in signal intensity from the first spectral range with saved profiles . This enables a more accurate moisture detection . Moist structures usually absorb more visible light or have a stronger spectrum in darker colors .
[0037] In another embodiment of the invention, the micro control unit and / or the connected processing unit cross-references the detection data of the second spectral range with saved material profiles and determines a matching profile by the signal and / or adds it to the saved profiles .
[0038] This allows for certain types of detection as well as feeding and improving an Al algorithm or machine learning unit as well as sharing information with other detection structures worldwide or within a cloud architecture and / or connected by other means such as ethernet , radio , or Bluetooth, to let other devices be improved in their delivered results and capabilities .
[0039] In another embodiment of the invention the detection method uses reference data from agriculture plants . This allows for applications on modern farming and agriculture were moisture and over time detection is key to deliver good yields . In another embodiment of the invention the optical sensor uses filters for 960 nm, 1460 nm, 1670 nm and / or 1960 nm . This covers typical water or vapor absorption bands indicative for moisture . This further allows a broad dataset of existing IR profiles and data to be used and also is commercially attractive because of compatibilities with other hardware and software .
[0040] Furthermore , a detection structure is provided, comprising an optical sensor and a micro control unit , wherein the sensor comprises spectral channels for a first spectral range and a second spectral range , and wherein the micro control unit is configured to execute the optical detection method according to the foregoing description .
[0041] Preferably, the detection structure comprises a broadband light source configured to emit light with spectral components of at least the first spectral range and the second spectral range . This allows for detection in the absence of other light sources and thus broadens the scope of service of the detection structure .
[0042] To summari ze , in slightly di f ferent words and with a di f ferent emphasis than before : Since color of an obj ect is often related to its moisture content , the present invention aims to combine spectral color sensing and moisture sensing into one product or sensing platform / structure . Therefore , spectral color and moisture measurements are preferably combined into one sensing device . Core of the invention is to combine a number of (preferably at least three ) spectral channels in the visible range with multiple (preferably at least three ) spectral channels in the infrared range all into one detector . In particular, the near-infrared (NIR) range is used, alternatively the short wavelength infrared ( SWIR) range . Hence , by combining spectral channels in the visible range and the infrared range , the moisture level of obj ects may be derived . In addition, or alternatively, di f ferent materials constituting an at-least partially reflective obj ect may be identi fied and distinguished from each other by employing said method . BRIEF DESCRIPTION OF THE DRAWINGS
[0043] Exemplary embodiments of the invention are subsequently discussed with reference to the accompanying drawings .
[0044] FIG . 1 shows an exemplary embodiment of an optical detection structure according to the invention .
[0045] FIG . 2 shows an exemplary graph of a broadband light spectrum and corresponding spectral sensitivities of individual optical channels implemented in the detection structure of FIG . 1 .
[0046] DETAILED DESCRIPTION
[0047] FIG . 1 schematically presents an embodiment of a detection structure 100 or a detection system according to the invention . The detection structure 100 implements a detection method based on reflective spectral measurements in at least two di f ferent spectral regions or ranges , in particular the visible range and the near-infrared (NIR) range , preferably combined in a single detection device or sensor . Corresponding data flows within the detection structure 100 are indicated by double-headed arrows in FIG . 1 .
[0048] The detection structure 100 in this embodiment comprises a broadband light source 2 ( alternatively, the light source 2 may be an external component ) . The broadband light source 2 emits light 4 with a broadband spectrum, preferably with notable contributions in the visible range and in the near-infrared range . In this case , even though not shown, the broadband light source 2 is preferably connected to the optical sensor 8 or the micro control unit 12 ( see below) to make it controllable by the detection structure 100 .
[0049] A measurement sample 500 or detection obj ect of interest reflects the light emitted by the light source 2 towards the optical sensor 8 ( in an alternative "transmissive" arrangement the light passes through a transmissive obj ect , i . e . light source 2 and optical sensor 8 are arranged roughly at opposite sides of the obj ect ) . The reflected portion of the light is designated the reference numeral 6 . The optical sensor 8 detects or senses raw spectral data 10 and passes them to the micro control unit (MCU) 12 .
[0050] In detail , the optical sensor 8 detects reflective spectral data from a first spectral range 20 , preferably the visible range , by three or more light filters and from a second spectral range 30 , preferably, the near-infrared range , by three or more light filters and sends the detection data to the micro control unit 12 . Preferably, there are even 3-5 , 6- 8 , or 6- 9 , for instance , spectral channels or filters for detection for each of the tow ranges .
[0051] The micro control unit 12 and the detection architecture allow for several options on processing the data . It could either be processed locally (within a local data host ) or, at least to some extent , in a remote or cloud architecture 16 . Independent of the processing locali zation, the data is preferably cross-referenced then to each other and / or further improved by processing results from one set of the detection data based on the other set of detection data . In more detail , it can be arranged that the micro control unit 12 or the connected ( e . g . , cloud-based) processing unit improves the processing results from the second spectral range 30 by processing the first spectral range 20 ( or vice versa ) and comparing / referencing / aligning the results with each other .
[0052] Subsequently, or alternatively without prior cross-referencing, the detection data of the second spectral range 30 can be processed and / or referenced to a database to detect a moisture level of the measurement sample 500 by the micro control unit 12 or connected processing unit ( s ) .
[0053] With this it is also possible to let data of the second spectral range 30 be cross-referenced by the data of the first spectral range 20 which is beforehand processed and / or referenced to a database to detect a color and / or a moisture level of the measurement sample 500 by the micro control unit 12 or the connected processing unit ( s ) . In this structure it is optional to let a machine learning and / or arti ficial intelligence algorithm be used by the micro control unit 12 and / or by the connected processing unit to cross reference and improve the detection data and save the improved result .
[0054] It is also possible within this detection structure 100 to let the detection method be applied over time and measure a change over time and / or store the data by the micro control unit 12 and / or the connected processing unit .
[0055] Additionally, it can be possible to let the connected processing unit and / or the micro control unit 12 correlate the change in signal intensity from the first spectral range 20 and / or the second spectral range 30 with saved data profiles .
[0056] The detection structure 100 allows the implemented detection method to optionally let the connected processing unit and / or the micro control unit 12 cross-reference the detection data of the second spectral range 30 with saved material profiles and determine material properties of the measurement sample 500 .
[0057] FIG . 2 shows a graph of a broadband spectrum 40 of light detected . In all the embodiments on the invention two spectral ranges are detected . In this graph the first spectral range 20 corresponds to visible light , and the second spectral range 30 corresponds to infrared light . Blue , green, and red color light intensity graphs 22 , 24 , 26 are shown exemplarily within the visible range of the complete spectrum . Similarly, di f ferent channels within the infrared range are indicated . The percentage of intensity ( corresponding to spectral sensitivity) is plotted over the wavelength in this graph . Each of the subunit graphs is detectable with the help of a corresponding light filter for the optical sensor 8 .
[0058] The present invention has various applications . These are based on the cognition that quite frequently the color of an obj ect is changing i f the moisture of the obj ect is changing . Notable examples are : • Moisture on floor: color of floor appears different when moisture is present (e.g., wet carpet appears darker than dry carpet) .
[0059] • Reduction of moisture in fruits: fruit color will change while moisture is degrading.
[0060] • Reduction of moisture in meat: meat color will change while moisture is degrading.
[0061] • Reduction of moisture in agri-food production: color is changing wile moisture is adding up or degrading (e.g., rice, soybeans, grain, coffee & cacao beans etc.) .
[0062] • Gras growing on well moisturized soil appearing different in color than if soil is not well moisturized.
[0063] Hence, as the examples indicate, color change and moisture change often are in a direct relationship to each other. Even where there is no 1:1 correspondence, a general or partial correlation can be used for machine learning concepts to also distinguish cross-relations and increase accuracy in moisture measurements.
[0064] To this end, the invention provides a sensing platform or sensing structure 100 with an optical sensor 8 and an associated micro control unit 12 which can measure - with a single sensor 8 - both the color of an object and the moisture level of an object. Both absolute and relative measurements may be done and provided to a host (the micro control unit 12 and / or a connected processing unit, in particular a cloud computing architecture 16) where data is post-processed for further decisions and / or control operations of the final application. This is done by combining a number of (preferably >=3) spectral visible filters (e.g., in the X-Y-Z color space) for color measurements and multiple (preferably >=3) spectral filters in the infrared range, in particular the near infrared (NIR) range, but possibly also / alternatively in the short wavelength infrared (SWIR) range from 1000 nm to 2200 nm. In addition of measuring moisture, the NIR spectral channels can also provide information about material type of the illuminated object in question (such as is the material a fabric, plastic, wooden floor, carpet, etc.) .
[0065] Advantages of such a combination of reflective object color and moisture measurement include :
[0066] • Customers or users can use one sensor to cover both use-cases ( color & moisture determination) .
[0067] • Smaller form factor possible
[0068] • Lower cost due to less bill of materials (BOM)
[0069] • Simpler implementation ( less complexity for light path design, etc . )
[0070] In general , more ( and possibly better ) data provided by the sensing platform allow for better decisions or operation of the final product or application . Returning to the previous examples , the concrete advantages may be as follows :
[0071] • Example fruits : Color of an apple can be measured, in addition the sensor can provide information on how long it is on the shel f (by measuring moisture content ) , or how sweet it really is .
[0072] • Example floor type cleaning : A robot cleaner can identi fy if the floor is wet or dry to adj ust a cleaning program . Color measurement is an additional input since wet floor appears darker than dry floor . Floor type detection is used to apply suitable amounts of water on suitable floor types for cleaning . Some floor types , however, like wooden floors should not be exposed to high amounts of water .
[0073] • Example soil / gras : A robot mower measures color of gras and moisture of soil . Based on these two parameters , a user can be informed about health of gras or soil . Answers to the following question may be given : Fertili zer needed? Moistening needed? I f gras is wet , a robot mower can stop to operate to lengthen li fetime of mower .
[0074] It will be appreciated that the disclosure is not limited to the disclosed embodiments and to what has been particularly shown and described hereinabove . Rather, features recited in separate dependent claims or in the description may advantageously be combined . Furthermore , the scope of the disclosure includes those variations and modi fications , which will be apparent to those skilled in the art and fall within the scope of the appended claims . The term "comprising", insofar it was used in the claims or in the description, does not exclude other elements or steps of a corresponding feature or procedure. In case that the terms "a" or "an" were used in conjunction with features, they do not exclude a plurality of such features. Moreover, any reference signs in the claims should not be construed as limiting the scope.
[0075] LIST OF REFERENCE SIGNS broadband light source emitted light reflected light optical sensor spectral data micro control unit exchanged pre-processed data cloud architecture first spectral range red light channel green light channel blue light channel second spectral range , 34 , 36 , 38 first to fourth infrared light channel broadband spectrum 0 detection structure 0 measurement sample or obj ect
Claims
CLAIMS1. An optical detection method based on detection and analysis of light reflected from an illuminated object (500) and / or of light passing through a transmissive object (500) , wherein a) an optical sensor (8) detects spectral data (10) from a first spectral range (20) and a second spectral range (30) , the second spectral range (30) being different from the first spectral range (20) , b) the sensor (8) sends the detected spectral data (10) to a micro control unit (12) which receives and processes the spectral data (10) from the optical sensor (8) , and wherein the micro control unit (12) and / or a connected processing unit derives a number of physical properties of the object (500) from said spectral data (10) .
2. The detection method according to claim 1, wherein the first spectral range (20) comprises wavelengths in the visible range, in particular from 300 nm to 690 nm, and the second spectral range (30) comprises wavelengths in the infrared range, in particular from 690 nm to 3000 nm.
3. The detection method according to any one of the preceding claims, wherein the second spectral range (30) comprises wavelengths in the near-infrared range, in particular from 690 to 1000 nm, or in the short wavelength infrared range, in particular from 1000 nm to 2200 nm.
4. The detection method according to according to any one of the preceding claims, wherein the micro control unit (12) and / or a connected processing unit cross-references and / or correlates the spectral data (10) from the first spectral range (20) and the second spectral range (30) to each other and thereby improves a detection result.
5. The detection method according to any one of the preceding claims, wherein the micro control unit (12) and / or the connected processing unit improves processing results obtained from thesecond spectral range (30) by processing the first spectral range (20) and referencing the results.
6. The detection method according to any one of the preceding claims, wherein the optical sensor (8) detects spectral data (10) from the first spectral range (20) by three or more light filters and spectral data (10) from the second spectral range (30) by three or more light filters.
7. The detection method according to any one of the preceding claims, wherein the spectral data (10) from the second spectral range (30) are used, in particular by referencing to a database, to detect a moisture level of the object (500) .
8. The detection method according to any one of the preceding claims, wherein the spectral data (10) from the first spectral range (20) are used, in particular by referencing to a database, to detect a color attribute of the object (500) .
9. The detection method according to any one of the preceding claims, wherein the spectral data (10) from the second spectral range (30) are cross-referenced to the spectral data (10) from the first spectral range (20) to detect a moisture level.
10. The detection method according to any one of the preceding claims, wherein the spectral data (10) from the second spectral range (30) are used, in particular by referencing to a database, to identify a material which constitutes the obj ect (500) or which the object (500) comprises.
11. The detection method according to any one of the preceding claims, wherein a machine learning algorithm and / or an artificial intelligence algorithm is used by the micro control unit (12) and / or by the connected processing unit to cross-reference the spectral data from the first spectral range (20) and the second spectral range (30) and thereby to improve a detection result.
12. The detection method according to any one of the preceding claims, wherein the detection is applied or executed over time,and a change over time is measured and stored by the micro control unit (12) and / or the connected processing unit.
13. The detection method according to any one of the preceding claims, wherein the optical sensor (8) uses light filters or optical channels for 960 nm, 1460 nm, 1670 nm and / or 1960 nm.
14. A detection structure (100) comprising an optical sensor (8) and a micro control unit (12) , wherein the sensor (8) comprises a plurality of spectral channels for a first spectral range (20) and a plurality of spectral channels for a second spectral range (30) , and wherein the micro control unit (12) and / or a connected processing unit is configured to execute the optical detection method according to any one of the preceding claims.
15. A detection structure (100) according to claim 13 which comprises a broadband light source (2) configured to emit light with spectral components of at least the first spectral range (20) and the second spectral range (30) .
16. Combined optical moisture and color sensor configured for using at least three spectral channels in the visible range for color detection and using at least three spectral channels in the infrared range for moisture detection and / or material identification, such as floor type identification.
Citation Information
Patent Citations
Spectroscopy having correction for broadband distortion for analyzing multi-component samples
US20100243900A1
Material analytical sensor and material analyzer
US20160169795A1
Spectrometry system applications
US20170292908A1
Apparatus and method for measuring and correlating characteristics of fruit with visible / near infra-red spectrum
US6512577B1
Imaging system for object recognition and assessment
US9959628B2
