CN112444493B - Optical detection system and device based on artificial intelligence - Google Patents
Optical detection system and device based on artificial intelligence Download PDFInfo
- Publication number
- CN112444493B CN112444493B CN202011088696.XA CN202011088696A CN112444493B CN 112444493 B CN112444493 B CN 112444493B CN 202011088696 A CN202011088696 A CN 202011088696A CN 112444493 B CN112444493 B CN 112444493B
- Authority
- CN
- China
- Prior art keywords
- model
- call
- optical detection
- terminal control
- preset
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 122
- 230000003287 optical effect Effects 0.000 title claims abstract description 81
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 31
- 238000001228 spectrum Methods 0.000 claims abstract description 53
- 239000000126 substance Substances 0.000 claims abstract description 37
- 238000012545 processing Methods 0.000 claims abstract description 27
- 238000004891 communication Methods 0.000 claims abstract description 16
- 239000013598 vector Substances 0.000 claims description 69
- 238000000034 method Methods 0.000 claims description 25
- 238000004458 analytical method Methods 0.000 claims description 20
- 239000000463 material Substances 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 8
- 230000003595 spectral effect Effects 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 abstract description 9
- 238000010183 spectrum analysis Methods 0.000 abstract description 5
- 239000000306 component Substances 0.000 description 11
- 230000009286 beneficial effect Effects 0.000 description 9
- 239000003814 drug Substances 0.000 description 8
- 238000003384 imaging method Methods 0.000 description 6
- 244000005700 microbiome Species 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000013527 convolutional neural network Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- WSFSSNUMVMOOMR-UHFFFAOYSA-N Formaldehyde Chemical compound O=C WSFSSNUMVMOOMR-UHFFFAOYSA-N 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 210000003811 finger Anatomy 0.000 description 3
- 235000013305 food Nutrition 0.000 description 3
- 230000004927 fusion Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000007635 classification algorithm Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 235000012041 food component Nutrition 0.000 description 2
- 239000005428 food component Substances 0.000 description 2
- 239000011521 glass Substances 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000000447 pesticide residue Substances 0.000 description 2
- 235000001674 Agaricus brunnescens Nutrition 0.000 description 1
- 208000034656 Contusions Diseases 0.000 description 1
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 description 1
- 240000009088 Fragaria x ananassa Species 0.000 description 1
- 241000233866 Fungi Species 0.000 description 1
- 244000068988 Glycine max Species 0.000 description 1
- 235000010469 Glycine max Nutrition 0.000 description 1
- 229920000877 Melamine resin Polymers 0.000 description 1
- 208000000112 Myalgia Diseases 0.000 description 1
- 208000031888 Mycoses Diseases 0.000 description 1
- 244000046052 Phaseolus vulgaris Species 0.000 description 1
- 235000010627 Phaseolus vulgaris Nutrition 0.000 description 1
- 241000533293 Sesbania emerus Species 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 208000034526 bruise Diseases 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 239000000796 flavoring agent Substances 0.000 description 1
- 235000019634 flavors Nutrition 0.000 description 1
- 235000015219 food category Nutrition 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 208000015181 infectious disease Diseases 0.000 description 1
- 230000001678 irradiating effect Effects 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 235000013372 meat Nutrition 0.000 description 1
- JDSHMPZPIAZGSV-UHFFFAOYSA-N melamine Chemical compound NC1=NC(N)=NC(N)=N1 JDSHMPZPIAZGSV-UHFFFAOYSA-N 0.000 description 1
- 230000002906 microbiologic effect Effects 0.000 description 1
- 208000013465 muscle pain Diseases 0.000 description 1
- 239000000843 powder Substances 0.000 description 1
- 238000002331 protein detection Methods 0.000 description 1
- 102000004169 proteins and genes Human genes 0.000 description 1
- 108090000623 proteins and genes Proteins 0.000 description 1
- 235000021251 pulses Nutrition 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 235000020095 red wine Nutrition 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 235000021012 strawberries Nutrition 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 210000003813 thumb Anatomy 0.000 description 1
- 229940126680 traditional chinese medicines Drugs 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 235000013311 vegetables Nutrition 0.000 description 1
- 238000001845 vibrational spectrum Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/3581—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using far infrared light; using Terahertz radiation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/06—Protocols specially adapted for file transfer, e.g. file transfer protocol [FTP]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Toxicology (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention provides an optical detection system and device based on artificial intelligence, wherein the system comprises: the optical detection device is used for shooting a substance spectrum image; the terminal control equipment is in communication connection with the optical detection device and is used for controlling the working mode of the optical detection device and acquiring a substance spectrum image shot by the optical detection device; the artificial intelligent cloud processing platform is in communication connection with the terminal control equipment and is used for acquiring a substance spectrum image transmitted by the terminal control equipment, processing the substance spectrum image by adopting a preset model, acquiring a detection report and transmitting the detection report to the terminal control equipment. The optical detection system based on artificial intelligence is based on a spectrum analysis technology, combines the technologies of artificial intelligence, big data, cloud computing, the Internet of things and the like, provides a rapid, convenient and easy-to-use intelligent detection product and a comprehensive solution for a user, simplifies the flow of substance detection, reduces the cost and enables the user to easily realize on-site rapid detection.
Description
Technical Field
The invention relates to the technical field of detection, in particular to an optical detection system and device based on artificial intelligence.
Background
At present, the spectrum is a pattern formed by sequentially arranging dispersed monochromatic light according to the wavelength after the light of the complex color is split by a dispersion system; the specific spectrum of the article implies information on the composition and content of the substance; the spectrum analysis is a method for identifying substances and determining chemical compositions and relative contents of the substances according to the spectra of the substances, but the existing spectrum analysis products have single functions and are unfavorable for the large volume of instruments and the field detection.
Disclosure of Invention
The invention aims to provide an optical detection system based on artificial intelligence, which is based on a spectrum analysis technology, combines the technologies of artificial intelligence, big data, cloud computing, the Internet of things and the like, provides a rapid, convenient and easy-to-use intelligent detection product and a comprehensive solution for a user, simplifies the flow of substance detection, reduces the cost and enables the user to easily realize on-site rapid detection.
The embodiment of the invention provides an optical detection system based on artificial intelligence, which comprises:
the optical detection device is used for shooting a substance spectrum image;
the terminal control equipment is in communication connection with the optical detection device and is used for controlling the working mode of the optical detection device and acquiring a substance spectrum image shot by the optical detection device;
the artificial intelligent cloud processing platform is in communication connection with the terminal control equipment and is used for acquiring a substance spectrum image transmitted by the terminal control equipment, processing the substance spectrum image by adopting a preset model, acquiring a detection report and transmitting the detection report to the terminal control equipment; the terminal control device displays the detection report.
Preferably, the optical detection device includes: hyperspectral cameras and/or terahertz spectrometers.
Preferably, the terminal control apparatus includes: one or more of a mobile phone, a tablet and a computer are combined.
Preferably, the preset model includes: one or more of a three-dimensional convolution model, a double-branch convolution model and a small sample convolution model.
Preferably, the artificial intelligence cloud processing platform performs the following operations:
before a preset model is adopted to process a material spectrum image, model calling information sent by terminal control equipment is received, and a preset model is obtained from a preset model calling library based on the model calling information;
wherein the model call library comprises: the system comprises a permission calling vector, model numbers corresponding to the permission calling vector one by one and preset models corresponding to the model numbers one by one; wherein the permission call vector is as follows:
X i ＝(x i1 ，x i2 ，…，x im )；
wherein X is i Is a permission call vector corresponding to the ith model number; a, a im The value of the mth call parameter in the ith permission call vector;
obtaining a preset model from a preset model call library based on model call information, wherein the method comprises the following steps:
analyzing model calling information based on a preset analysis template to obtain a model calling vector; the model call vector is as follows:
Y＝(y 1 ，y 2 ，…，y m )；
wherein Y represents a model call vector; y is m Representing the value of the mth call parameter in the model call vector; in the analysis, when the values of the calling parameters of the analysis template are not analyzed from the model calling information, filling the values of the calling parameters by adopting preset filling values;
the first similarity between the model call vector and each license call vector is calculated as follows:
wherein Sim (Y, X p ) Representing a first similarity between the model call vector Y and the p-th permission call vector; y is q Representing the value of the q-th call parameter in the model call vector; x is x pq A value of a q-th call parameter representing a p-th permission call vector;
when the maximum value in all the values of the first similarity is larger than or equal to a preset first threshold value and smaller than a second preset threshold value, obtaining a model number corresponding to a permission calling vector of the maximum value of the first similarity, and calling a preset model corresponding to the model number one by one based on the model number; when the maximum value of all the first similarity values is greater than or equal to a second preset threshold value, extracting all the model numbers corresponding to the permission calling vectors with the first similarity values greater than the second preset threshold value, obtaining preset model description information corresponding to the model numbers, preparing a to-be-selected list from the model description information and the model numbers, sending the to-be-selected list to terminal control equipment, receiving selection operation of the terminal control equipment on the to-be-selected list, analyzing the selection operation, obtaining the model numbers selected and called by a user, and calling preset models corresponding to the model numbers one by one based on the model numbers; wherein the selecting operation includes multiple selections; the first threshold value is smaller than a second preset threshold value;
when the maximum value of all the first similarity values is smaller than a preset first threshold value and/or the number of the calling parameter values filled by adopting a filling means in the analysis process is larger than the preset number, a history calling record of the terminal control equipment is obtained, a temporary calling library is built based on the history calling record, the second similarity of a model calling vector and a permission calling vector in the temporary calling library is calculated, and the calculation formula is as follows:
wherein Sim (Y, L j ) A second similarity between the model call vector Y and a jth permission call vector in the temporary call library; x is x jq A value representing a qth call parameter of a jth allowable call vector in the temporary call library;
calling a model corresponding to the maximum value of the second similarity;
the calling parameters include: the voltage and current of the optical detection device, the spectrum wavelength of the spectrum image of the shooting substance, the light intensity, the focal length of the lens and the depth of field of the lens.
Preferably, the optical detection system based on artificial intelligence further comprises: the model calling two-dimensional code comprises: model calling information and/or setting information of the optical detection device;
when a key of the optical detection device is pressed for a long time to reach a preset time, the optical detection device enters a model calling and setting mode, and when the model calling and setting mode is adopted, the optical detection device shoots a model calling two-dimensional code;
the terminal control equipment acquires a model calling two-dimensional code through the optical detection device, and acquires model calling information and/or setting information based on the model calling two-dimensional code;
the terminal control equipment sets shooting parameters of the optical detection device based on the setting information;
the terminal control equipment sends the model calling information to an artificial intelligent cloud processing platform, and the artificial intelligent cloud processing platform calls the model based on the model calling information; the model call information includes: model number of the model.
Preferably, the artificial intelligence cloud processing platform further performs the following operations:
receiving model calling information sent by a terminal control device, and sending a model corresponding to the model calling information to the terminal control device when the model calling information is the same as the model calling information of the previous N times; the terminal control device saves the model.
The invention also provides an optical detection device based on artificial intelligence, which comprises:
the shell body is provided with a plurality of grooves,
a shooting window arranged at one end of the shell,
the display screen is arranged at the other end of the shell;
the key is arranged on one side of the shell, and a hand holding line which is suitable for four fingers of a human hand is arranged on one side of the shell away from the key;
the shooting module is arranged in the shell and is used for shooting a spectrum image of a substance;
the controller is arranged in the shell and is electrically connected with the shooting module, the display screen and the keys respectively;
and the wireless communication module is electrically connected with the controller and is used for being in communication connection with the terminal control equipment.
Preferably, the photographing module includes:
the first lens assembly is sleeved with a first gear at the periphery thereof;
at least one second lens component, wherein a second gear is sleeved on the periphery of the second lens component;
the first gear and the second gear are arranged in the inner gear ring, the first gear is meshed with the second gear, and the second gear is meshed with the inner teeth of the inner gear ring;
one end of a stator of the rotating shaft is fixedly connected with the shell, and the other end of the stator of the rotating shaft is fixedly connected with the first lens component;
the first connecting rods are respectively and vertically arranged with the central axis of the rotating shaft and are respectively and fixedly connected with the rotor of the rotating shaft;
the second connecting rods are perpendicular to the first connecting rods and parallel to the central axis of the rotating shaft; the second connecting rod and the first connecting rod are in one-to-one correspondence with the second lens component; one end of the second connecting rod is rotationally connected with one end of the first connecting rod far away from the rotating shaft, and the other end of the second connecting rod is fixedly connected with the middle part of the U-shaped fixing piece; both ends of the U-shaped fixing piece are fixedly connected with one side of the second gear.
Preferably, the second lens assembly includes:
the body is provided with a plurality of grooves,
the annular body is sleeved on the outer periphery of the body, and two first rotating bodies are symmetrically arranged between the outer periphery of the body and the inner periphery of the annular body; the rotating end of the first rotating body is fixedly connected with the body; the fixed end of the first rotating body is fixedly connected with the annular body;
two second rotating bodies which are symmetrically arranged and are arranged between the annular body and the second gear; the fixed end of the second rotating body is fixedly connected with the periphery of the annular body; the rotating end of the second rotating body is fixedly connected with the inner periphery of the second gear;
the central axes of the two first rotating bodies are perpendicular to the central axes of the two second rotating bodies.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:
FIG. 1 is a schematic diagram of an optical detection system based on artificial intelligence in an embodiment of the invention;
FIG. 2 is a method of object-level spectral modeling based on a three-dimensional convolutional neural network;
FIG. 3 is a method of modeling hyperspectral data based on a multi-branch convolutional network;
FIG. 4 is a small sample fast adaptation hyperspectral modeling method based on meta-learning;
FIG. 5 is a schematic diagram of an optical detection device according to an embodiment of the present invention;
fig. 6 is a schematic diagram of positions of an inner gear ring, a second gear and a first gear of an optical detection device according to an embodiment of the present invention.
In the figure:
1. an optical detection device; 2. a terminal control device; 3. an artificial intelligent cloud processing platform; 11. a housing; 12. a shooting window; 13. a display screen; 14. a key; 15. a controller; 16. a wireless communication module; 17. a second gear; 18. a first gear; 19. an inner gear ring; 4. a shooting module; 41. a rotating shaft; 42. a first link; 43. a second link; 44. a first lens assembly; 45. a U-shaped fixing piece; 46. a second lens assembly; 47. an annular body; 48. a first rotating body; 49. a second rotating body; 50. a body.
Detailed Description
The preferred embodiments of the present invention will be described below with reference to the accompanying drawings, it being understood that the preferred embodiments described herein are for illustration and explanation of the present invention only, and are not intended to limit the present invention.
An embodiment of the present invention provides an optical detection system based on artificial intelligence, as shown in fig. 1, including: an optical detection device 1 for photographing a spectrum image of a substance;
the terminal control equipment 2 is in communication connection with the optical detection device 1 and is used for controlling the working mode of the optical detection device 1 and acquiring a substance spectrum image shot by the optical detection device 1;
the artificial intelligent cloud processing platform 3 is in communication connection with the terminal control equipment 2 and is used for acquiring a substance spectrum image transmitted by the terminal control equipment 2, processing the substance spectrum image by adopting a preset model, acquiring a detection report and transmitting the detection report to the terminal control equipment 2; the terminal control device 2 displays the detection report.
The working principle and the beneficial effects of the technical scheme are as follows:
a user uses the optical detection device 1 to shoot a substance spectrum image of a detected object; the material spectrum image is uploaded to the artificial intelligent cloud processing platform 3 through the terminal control equipment 2, the artificial intelligent cloud processing platform 3 processes the material spectrum image by adopting a preset model, the optical detection device 1 only takes charge of collecting the material spectrum image, so that the volume of the material spectrum image can be reduced as much as possible, and when the material spectrum image is applied to real-time detection on a flow line, light sources can be erected before and after the optical detection device 1 to irradiate the region shot by the optical detection device 1; the artificial intelligent cloud processing platform 3 stores models corresponding to various detections, so that the diversification of the detection types is realized, and the diversification of application scenes is realized; the application scene comprises: analysis of the component content of the industrial screen, such as formaldehyde content detection; industrial class identification, such as plastic class identification; analysis of the content of agricultural products, such as protein detection, sugar detection, moisture detection and the like; agricultural product variety identification, for example: identification between arabica and robusta beans; food component content analysis such as apple sweetness analysis, alcohol content analysis, melamine detection, etc.; food category identification, such as white spirit flavor identification, red wine quality identification and the like; analyzing the content of the components of the medicine, and detecting whether the medicine is compliant; drug class identification, realizing rapid classification; chinese medicine adulteration identification, such as rhizoma Pinelliae adulterated with rhizoma arisaematis, radix Angelicae Pubescentis adulterated with radix Angelicae sinensis, etc.; judicial government affair authentication, such as handwriting ink authentication, etc.; detecting purity of the articles, such as flaw detection of ornaments, glass and other substances; and detecting the purity of the object, such as flaw detection of ornaments, glass and the like. The method comprises the steps of irradiating food by adopting a spectrum in advance, enabling a machine to learn deeply by using the spectrum data and the detection result of the food which are fed back, forming a preset model, and then using the preset model for detection; detecting food components, pesticide residues, sweetness and acidity. Judging whether the medicine concentrated in the traditional Chinese medicine is true or false or not, and pesticide residues of agricultural products.
The optical detection system based on artificial intelligence is based on a spectrum analysis technology, combines the technologies of artificial intelligence, big data, cloud computing, the Internet of things and the like, provides a rapid, convenient and easy-to-use intelligent detection product and a comprehensive solution for a user, simplifies the flow of substance detection, reduces the cost and enables the user to easily realize on-site rapid detection.
In one embodiment, the optical detection device 1 comprises: hyperspectral cameras and/or terahertz spectrometers.
The working principle and the beneficial effects of the technical scheme are as follows:
the hyperspectral camera is used for shooting hyperspectral images, and can obtain the spatial morphology information of the measured object and the spectrum information of each pixel point to form a three-dimensional data block. Thus, the on-line nondestructive detection of the components of the detected object can be realized. The method is applied to detection of the effective components of western medicine tablets; detecting varieties of coffee beans and soybeans; identifying adulteration of traditional Chinese medicine; and detecting the bruise of the strawberries. Specific application examples are as follows: the TVC (total plate counting) and the plate counting of meat, aquatic products, vegetables, mushrooms and other products can be rapidly and nondestructively detected by adopting visible and near infrared band hyperspectral sensing and combining various chemometrics and machine learning algorithms such as PLS (partial least squares) PLS model), MLR (multiple linear regression), SVM (support vector machine model), ANN (artificial neural network). The method adopts visible and near infrared band hyperspectral sensing and combines with various machine learning classification algorithms such as SVM, LS-SVM, ANN and the like, can realize fungus infection classification of various foods and agricultural products, and can realize visualization of infection parts. The fungus variety classification can be realized by adopting visible and near infrared band hyperspectral sensing and combining with SVM, KNN, ANN and other machine learning classification algorithms; meanwhile, by adopting means such as PCA and the like, analysis and detection of the mycelium growth process can be realized.
Terahertz waves refer to electromagnetic waves of 0.1THz-10THz, have the characteristics of light waves and microwaves, and have the advantages of safety, perspective and spectrum resolution. According to the imaging principle, the terahertz imaging technology can be largely classified into a continuous terahertz wave and a pulsed terahertz wave imaging technology. The imaging based on the pulse wave can provide richer and more accurate information, and the continuous terahertz wave imaging has the advantages of simple equipment and high imaging speed. Terahertz waves can excite molecular motion of low-frequency biomolecules such as proteins and DNA, so that microbiological structural and kinetic information which cannot be obtained by other vibration spectrums can be obtained. Research shows that different types of microorganisms show different time-domain terahertz spectrum characteristics, so that the method can be used for classifying and identifying microorganisms and has certain potential in the field of new species discovery. In addition, the differences of the terahertz wave absorptivity of the microorganisms in different living states, death and powder are obvious. The research shows that the microorganism concentration has strong correlation with the terahertz resonance frequency shift, so that terahertz sensing can be used for quantitatively detecting the microorganism concentration. At present, research in the field focuses on improving detection sensitivity, for example, designing a sensor based on terahertz metamaterial, and realizing quantitative detection of micro microorganisms.
In one embodiment, the terminal control device 2 includes: one or more of a mobile phone, a tablet and a computer are combined.
In one embodiment, the predetermined model includes: one or more of a three-dimensional convolution model, a double-branch convolution model and a small sample convolution model.
The working principle and the beneficial effects of the technical scheme are as follows:
three-dimensional convolution model: as shown in fig. 2, in order to realize automatic object-level high-spectrum data classification based on a deep learning technology, a watershed algorithm is adopted to perform image segmentation and sample data standardization in the object-level high-spectrum modeling method based on a three-dimensional convolutional neural network; and sending the standardized data into a three-dimensional convolutional neural network, and simultaneously realizing extraction and fusion of the depth morphology features and the spectrum features.
Double convolution model: as shown in fig. 3, in the hyperspectral data modeling method based on the multi-branch convolutional network, in order to simplify model parameters and improve model robustness, morphology and spectral features are extracted by adopting the double-branch convolutional neural network respectively, and finally feature fusion is performed. The extraction of spectral features and spatial features and the deep fusion of the two features are realized through the structural design of the neural network.
Small sample convolution model: as shown in fig. 4, the small sample rapid adaptation hyperspectral modeling method based on meta-learning adopts a meta-learning method, and parallelizes and pre-trains the model by using the public hyperspectral dataset. The method is tested on a plurality of data sets of agricultural products, traditional Chinese medicines and other products, and results show that the accuracy and the robustness of small sample modeling can be remarkably improved.
In one embodiment, the artificial intelligence cloud processing platform 3 performs the following operations:
before a preset model is adopted to process a material spectrum image, model calling information sent by a terminal control device 2 is received, and a preset model is obtained from a preset model calling library based on the model calling information;
wherein the model call library comprises: the system comprises a permission calling vector, model numbers corresponding to the permission calling vector one by one and preset models corresponding to the model numbers one by one; wherein the permission call vector is as follows:
X i ＝(x i1 ，x i2 ，…，x im )；
wherein X is i Is a permission call vector corresponding to the ith model number; a, a im The value of the mth call parameter in the ith permission call vector;
obtaining a preset model from a preset model call library based on model call information, wherein the method comprises the following steps:
analyzing model calling information based on a preset analysis template to obtain a model calling vector; the model call vector is as follows:
Y＝(y 1 ，y 2 ，…，y m )；
wherein Y represents a model call vector; y is m Representing the value of the mth call parameter in the model call vector; in the analysis, when the values of the calling parameters of the analysis template are not analyzed from the model calling information, filling the values of the calling parameters by adopting preset filling values;
the first similarity between the model call vector and each license call vector is calculated as follows:
wherein Sim (Y, X p ) Representing a first similarity between the model call vector Y and the p-th permission call vector; y is q Representing the value of the q-th call parameter in the model call vector; x is x pq A value of a q-th call parameter representing a p-th permission call vector;
when the maximum value in all the values of the first similarity is larger than or equal to a preset first threshold value and smaller than a second preset threshold value, obtaining a model number corresponding to a permission calling vector of the maximum value of the first similarity, and calling a preset model corresponding to the model number one by one based on the model number; when the maximum value of all the first similarity values is greater than or equal to a second preset threshold value, extracting all the model numbers corresponding to the permission calling vectors with the first similarity values greater than the second preset threshold value, obtaining preset model description information corresponding to the model numbers, preparing a to-be-selected list from the model description information and the model numbers, sending the to-be-selected list to the terminal control equipment 2, receiving the selection operation of the terminal control equipment 2 on the to-be-selected list, analyzing the selection operation, obtaining the model numbers selected and called by a user, and calling preset models corresponding to the model numbers one by one based on the model numbers; wherein the selecting operation includes multiple selections; the first threshold value is smaller than a second preset threshold value;
when the maximum value of all the values of the first similarity is smaller than a preset first threshold value and/or the number of the values of the call parameters filled by adopting the filling means in the analysis process is larger than the preset number, a history call record of the terminal control device 2 is obtained, a temporary call library is built based on the history call record, the second similarity of the model call vector and the permission call vector in the temporary call library is calculated, and the calculation formula is as follows:
wherein Sim (Y, L j ) A second similarity between the model call vector Y and a jth permission call vector in the temporary call library; x is x jq A value representing a qth call parameter of a jth allowable call vector in the temporary call library;
calling a model corresponding to the maximum value of the second similarity;
the calling parameters include: the voltage, current, spectrum wavelength of the spectrum image of the shooting substance, light intensity, lens focal length and lens depth of field of the optical detection device 1.
The working principle and the beneficial effects of the technical scheme are as follows:
the artificial intelligent cloud processing platform 3 analyzes model call vectors from model call information sent by the terminal control equipment 2, matches the model call vectors with permission call vectors corresponding to all models in a model call library, determines the model required to be called by the terminal control equipment 2, analyzes a material spectrum image by the called model, and therefore the accuracy of the called model is directly related to the accuracy of a final analysis result; the model calling information sent to the artificial intelligent cloud processing platform 3 by the terminal control equipment 2 can be manually input by a user, or can directly take the existing state of the optical detection device 1 as the model calling information; the states of the optical detection device 1 include: the voltage, the current, the spectrum wavelength, the light intensity, the lens focal length, the lens depth of field and the like of the optical detection device 1; the method is characterized in that corresponding models are respectively obtained aiming at various different states, the recognition rate of the models is improved, and the accuracy of a final detection report is improved; when the model calling information is insufficient to select the called model, the intention model of the terminal control device 2 is intelligently judged according to the calling record of the terminal control device 2.
In one embodiment, the artificial intelligence based optical detection system further comprises: the model calling two-dimensional code comprises: model call information and/or setting information of the optical detection device 1;
when the key 14 of the optical detection device 1 is pressed for a long time to reach a preset time, the optical detection device 1 enters a model calling and setting mode, and when the model calling and setting mode is adopted, the optical detection device 1 shoots a model calling two-dimensional code;
the terminal control equipment 2 acquires a model calling two-dimensional code through the optical detection device 1, and acquires model calling information and/or setting information based on the model calling two-dimensional code;
the terminal control device 2 sets shooting parameters of the optical detection apparatus 1 based on the setting information;
the terminal control equipment 2 sends model calling information to the artificial intelligent cloud processing platform 3, and the artificial intelligent cloud processing platform 3 calls a model based on the model calling information; the model call information includes: model number of the model.
The working principle and the beneficial effects of the technical scheme are as follows:
when the optical detection device is used on a production line, when a substance to be detected is replaced, a professional is required to debug and set parameters of the optical detection device 1; the two-dimension code is called by the model, namely the needed model and setting parameters on the production line are printed on an adjusting card, when the production line is replaced to generate objects, the production line operator only needs to brush the two-dimension code of the needed model on the optical detection device 1, the terminal control equipment 2 reads the model to call the two-dimension code to realize automatic parameter setting of the optical detection device 1 and replacement of the corresponding calling model, and the operator can also quickly adjust according to differences of detection substances.
In one embodiment, the artificial intelligence cloud processing platform 3 also performs the following operations:
receiving model call information sent by the terminal control equipment 2, and sending a model corresponding to the model call information to the terminal control equipment 2 when the model call information is the same as the model call information of the previous N times; the terminal control apparatus 2 saves the model.
The working principle and the beneficial effects of the technical scheme are as follows:
when the same terminal control device 2 continuously calls the same model, the model is directly put on the front end [ terminal control device 2 ]; thus, the detection report can be completed directly at the terminal control device 2, and the detection rate is improved.
The invention also provides an optical detection device 1 based on artificial intelligence, as shown in fig. 5, comprising:
the housing 11 is provided with a plurality of openings,
a photographing window 12 provided at one end of the housing 11,
a display screen 13 provided at the other end of the housing 11;
the key 14 is arranged on one side of the shell 11, and a hand holding grain which is suitable for four fingers of a human hand is arranged on one side of the shell 11 away from the key 14;
a photographing module 4 disposed in the housing 11 for photographing a spectrum image of a substance;
the controller 15 is arranged in the shell 11 and is electrically connected with the shooting module 4, the display screen 13 and the keys 14 respectively;
the wireless communication module 16 is electrically connected with the controller 15 and is used for being in communication connection with the terminal control device 2.
The working principle and the beneficial effects of the technical scheme are as follows:
when the user uses the hand, the thumb of the hand is placed on the key 14, and the other four fingers hold the hand-held lines; the hand muscle pain can not be caused when the user holds the grains for a long time; the controller 15 controls the shooting module 4 to shoot a substance spectrum image, and sends the substance spectrum image to the terminal control equipment 2 through the wireless communication module 16, and the substance spectrum image is uploaded to the artificial intelligent cloud processing platform 3 through the terminal control equipment 2; the display screen 13 can display the state of the optical detection device 1, and a preset display interface is adopted for displaying when the optical detection device is in a model calling and setting mode; the shooting window 12 is used for shooting by the shooting module 4; a lens may be provided at the photographing window 12 for protecting the photographing module 4; in addition, a filter lens may be used to remove spectral signals that affect the analysis. The embodiment provides the optical detection device 1 which is convenient for users to hold, realizes the miniaturization and portability of detection equipment, and is convenient for inspection staff to take inspection and go out for use.
In one embodiment, as shown in fig. 5 and 6, the photographing module 4 includes:
a first lens assembly 44, wherein a first gear 18 is sleeved on the periphery of the first lens assembly 44;
at least one second lens assembly 46, the second gear 17 is sleeved on the periphery of the second lens assembly 46;
an inner gear ring 19, the first gear 18 and the second gear 17 are both arranged in the inner gear ring 19, the first gear 18 is meshed with the second gear 17, and the second gear 17 is meshed with the inner teeth of the inner gear ring 19;
a rotation shaft 41, one end of a stator of the rotation shaft 41 is fixedly connected with the housing 11, and the other end of the stator of the rotation shaft 41 is fixedly connected with the first lens assembly 44;
the first connecting rods 42 are respectively and vertically arranged with the central axis of the rotating shaft 41 and are respectively and fixedly connected with the rotor of the rotating shaft 41;
a plurality of second links 43 disposed perpendicular to the first links 42 and parallel to the central axis of the rotation shaft 41; the second connecting rod 43, the first connecting rod 42 and the second lens assembly 46 are in one-to-one correspondence; one end of the second connecting rod 43 is rotationally connected with one end of the first connecting rod 42 far away from the rotating shaft 41, and the other end is fixedly connected with the middle part of the U-shaped fixing piece 45; both ends of the U-shaped fixing member 45 are fixedly connected with one side of the second gear 17.
The working principle and the beneficial effects of the technical scheme are as follows:
the first lens assembly 44 is a main shooting component, the second lens assembly 46 is auxiliary shooting, three-dimensional modeling is carried out on a shooting object by combining pictures shot by the first lens assembly 44 and the second lens assembly 46, three-dimensional spectrum image acquisition is achieved, the second lens assembly 46 rotates in the inner toothed ring 19 under the cooperation of the rotating shaft 41, the first connecting rod 42, the second connecting rod 43 and the U-shaped fixing piece 45, the transformation angle is achieved, images of objects with multiple angles are provided for three-dimensional modeling, and the three-dimensional modeling is closer to the actual situation. In addition, the second lens component can be replaced by a spectrum generator for emitting a light source to irradiate the object, so that the first lens component can collect reflected light of the object to form a spectrum image.
In one embodiment, as shown in fig. 6, the second lens assembly 46 includes:
the body 50 is provided with a plurality of grooves,
the annular body 47 is sleeved on the outer periphery of the body 50, and two first rotating bodies 48 are symmetrically arranged between the outer periphery of the body 50 and the inner periphery of the annular body 47; the rotating end of the first rotating body 48 is fixedly connected with the body 50; the fixed end of the first rotating body 48 is fixedly connected with the annular body 47;
two second rotating bodies 49 symmetrically arranged, arranged between the annular body 47 and the second gear 17; the fixed end of the second rotating body 49 is fixedly connected with the outer periphery of the annular body 47; the rotating end of the second rotating body 49 is fixedly connected with the inner periphery of the second gear 17;
the central axes of the two first rotating bodies 48 are perpendicular to the central axes of the two second rotating bodies 49.
The working principle and the beneficial effects of the technical scheme are as follows:
the shooting angles of the second lens assembly 46 are adjusted through the first rotating body 48 and the second rotating body 49, so that images of objects shot from multiple angles are shot when the second lens assembly 46 is positioned in the same direction of the first lens assembly 44, the image basis in three-dimensional modeling is further enriched, and the accuracy of three-dimensional spectrum images after three-dimensional modeling is improved; the main body 50 is an optical unit for photographing of the second lens assembly 46.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (8)
1. An artificial intelligence based optical detection system, comprising:
an optical detection device (1) for capturing a spectral image of a substance;
the terminal control equipment (2) is in communication connection with the optical detection device (1) and is used for controlling the working mode of the optical detection device (1) and acquiring the substance spectrum image shot by the optical detection device (1);
the artificial intelligent cloud processing platform (3) is in communication connection with the terminal control equipment (2) and is used for acquiring the substance spectrum image transmitted by the terminal control equipment (2), processing the substance spectrum image by adopting a preset model, acquiring a detection report and transmitting the detection report to the terminal control equipment (2); the terminal control device (2) displays the detection report;
the preset model comprises the following steps: one or more of a three-dimensional convolution model, a double-branch convolution model and a small sample convolution model are combined;
the artificial intelligence cloud processing platform (3) performs the following operations:
before a preset model is adopted to process the material spectrum image, model calling information sent by the terminal control equipment (2) is received, and the preset model is obtained from a preset model calling library based on the model calling information;
wherein the model call library comprises: the system comprises a permission calling vector, model numbers corresponding to the permission calling vectors one by one and preset models corresponding to the model numbers one by one; wherein the permission call vector is as follows:
X i ＝(x i1 ，x i2 ，…，x im )；
wherein X is i Is the license call vector corresponding to the ith model number; x is X im The value of the mth call parameter in the ith permission call vector is set;
the obtaining the preset model from a preset model call library based on the model call information comprises the following steps:
analyzing the model call information based on a preset analysis template to obtain a model call vector; the model call vector is as follows:
Y＝(y 1 ，y 2 ，…，y m )；
wherein Y represents the model call vector; y is m A value representing an mth call parameter in the model call vector; in the analysis, when the value of the calling parameter of the analysis template is not analyzed from the model calling information, filling the value of the calling parameter by adopting a preset filling value;
calculating a first similarity between the model call vector and each of the license call vectors, wherein the calculation formula is as follows:
wherein,Sim(Y，X p ) Representing a first similarity between the model call vector Y and the p-th said permission call vector; y is q A value representing a qth call parameter in the model call vector; x is x pq A value representing a qth call parameter of a p-th said permission call vector;
when the maximum value in all the values of the first similarity is larger than or equal to a preset first threshold value and smaller than a second preset threshold value, acquiring the model number corresponding to the permission calling vector of the maximum value of the first similarity, and calling a preset model corresponding to the model number one by one based on the model number; when the maximum value of all the values of the first similarity is larger than or equal to a second preset threshold value, extracting all the model numbers corresponding to the permission call vectors with the first similarity larger than the second preset threshold value, obtaining preset model description information corresponding to the model numbers, preparing a list to be selected from the model description information and the model numbers, sending the list to be selected to the terminal control equipment (2), receiving the selection operation of the terminal control equipment (2) on the list to be selected, analyzing the selection operation, obtaining the model numbers selected and called by a user, and calling preset models corresponding to the model numbers one by one based on the model numbers; wherein the selecting operation includes multiple selections; the first threshold value is smaller than the second preset threshold value;
when the maximum value of all the values of the first similarity is smaller than a preset first threshold value and/or the number of the values of the call parameters filled by adopting a filling means in the analysis process is larger than a preset number, a history call record of the terminal control equipment (2) is obtained, a temporary call library is built based on the history call record, the second similarity of the model call vector and the permission call vector in the temporary call library is calculated, and the calculation formula is as follows:
wherein, sim (Y,L j ) A second similarity between the model call vector Y and the jth permission call vector in the temporary call library; x is x jq A value representing a qth one of said call parameters of a jth one of said permissible call vectors within said temporary call library;
invoking the model corresponding to the maximum value of the second similarity;
the call parameters include: the optical detection device (1) comprises voltage and current, and spectrum wavelength, light intensity, lens focal length and lens depth of field for shooting the spectrum image of the substance.
2. An artificial intelligence based optical detection system according to claim 1, characterized in that the optical detection device (1) comprises: hyperspectral cameras and/or terahertz spectrometers.
3. An artificial intelligence based optical detection system according to claim 1, characterized in that the terminal control device (2) comprises: one or more of a mobile phone, a tablet and a computer are combined.
4. The artificial intelligence based optical detection system of claim 1, further comprising: the method comprises the steps that a model calls a two-dimensional code, wherein the two-dimensional code is called by the model, and the two-dimensional code comprises: model call information and/or setting information of the optical detection device (1);
the key (14) of the optical detection device (1) is pressed for a long time to reach a preset time, the optical detection device (1) enters a model calling and setting mode, and when the model calling and setting mode is adopted, the optical detection device (1) shoots the model calling two-dimensional code;
the terminal control equipment (2) acquires the model calling two-dimensional code through the optical detection device (1), and acquires the model calling information and/or the setting information based on the model calling two-dimensional code;
the terminal control device (2) sets shooting parameters of the optical detection device (1) based on the setting information;
the terminal control equipment (2) sends the model calling information to the artificial intelligent cloud processing platform (3), and the artificial intelligent cloud processing platform (3) calls the model based on the model calling information; the model call information includes: model number of the model.
5. The artificial intelligence-based optical detection system of claim 1, wherein the artificial intelligence cloud processing platform (3) further performs the following operations:
receiving model call information sent by the terminal control equipment (2), and sending a model corresponding to the model call information to the terminal control equipment (2) when the model call information is the same as the model call information of the previous N times; the terminal control device (2) saves the model.
6. An artificial intelligence based optical detection system according to claim 1, characterized in that the optical detection device (1) comprises:
a housing (11),
a shooting window (12) arranged at one end of the shell (11),
the display screen (13) is arranged at the other end of the shell (11);
the key (14) is arranged on one side of the shell (11), and a hand holding line which is suitable for four fingers of a human hand is arranged on one side of the shell (11) away from the key (14);
the shooting module (4) is arranged in the shell (11) and is used for shooting a substance spectrum image;
the controller (15) is arranged in the shell (11) and is electrically connected with the shooting module (4), the display screen (13) and the keys (14) respectively;
and the wireless communication module (16) is electrically connected with the controller (15) and is used for being in communication connection with the terminal control equipment (2).
7. The artificial intelligence-based optical detection system of claim 6, wherein the camera module (4) comprises:
a first lens assembly (44), wherein a first gear (18) is sleeved on the periphery of the first lens assembly (44);
at least one second lens assembly (46), wherein a second gear (17) is sleeved on the periphery of the second lens assembly (46);
an inner gear ring (19), wherein the first gear (18) and the second gear (17) are both arranged in the inner gear ring (19), the first gear (18) is meshed with the second gear (17), and the second gear (17) is meshed with the inner teeth of the inner gear ring (19);
a rotating shaft (41), wherein one end of a stator of the rotating shaft (41) is fixedly connected with the shell (11), and the other end of the stator of the rotating shaft (41) is fixedly connected with the first lens assembly (44);
the first connecting rods (42) are respectively and vertically arranged with the central axis of the rotating shaft (41) and are respectively and fixedly connected with the rotor of the rotating shaft (41);
a plurality of second links (43) arranged perpendicularly to the first links (42) and parallel to the central axis of the rotation shaft (41); the second connecting rods (43), the first connecting rods (42) and the second lens components (46) are in one-to-one correspondence; one end of the second connecting rod (43) is rotationally connected with one end of the first connecting rod (42) which is far away from the rotating shaft (41), and the other end of the second connecting rod is fixedly connected with the middle part of the U-shaped fixing piece (45); both ends of the U-shaped fixing piece (45) are fixedly connected with one side of the second gear (17).
8. The artificial intelligence-based optical detection system of claim 7, wherein the second lens assembly (46) comprises:
a body (50),
an annular body (47) sleeved on the outer periphery of the body (50), wherein two first rotating bodies (48) are symmetrically arranged between the outer periphery of the body (50) and the inner periphery of the annular body (47); the rotating end of the first rotating body (48) is fixedly connected with the body (50); the fixed end of the first rotating body (48) is fixedly connected with the annular body (47);
two second rotating bodies (49) symmetrically arranged and arranged between the annular body (47) and the second gear (17); the fixed end of the second rotating body (49) is fixedly connected with the periphery of the annular body (47); the rotating end of the second rotating body (49) is fixedly connected with the inner periphery of the second gear (17);
the central axes of the two first rotating bodies (48) are perpendicular to the central axes of the two second rotating bodies (49).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011088696.XA CN112444493B (en) | 2020-10-13 | 2020-10-13 | Optical detection system and device based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011088696.XA CN112444493B (en) | 2020-10-13 | 2020-10-13 | Optical detection system and device based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112444493A CN112444493A (en) | 2021-03-05 |
| CN112444493B true CN112444493B (en) | 2024-01-09 |
Family
ID=74735979
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011088696.XA Active CN112444493B (en) | 2020-10-13 | 2020-10-13 | Optical detection system and device based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112444493B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113739039B (en) * | 2021-09-14 | 2023-04-07 | 中科蓝山怡源(北京)人工智能技术有限公司 | Inspection robot based on edge calculation |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1995987A (en) * | 2007-02-08 | 2007-07-11 | 江苏大学 | Non-destructive detection method and device for agricultural and animal products based on hyperspectral image technology |
| CN109341869A (en) * | 2018-12-04 | 2019-02-15 | 西安思科赛德电子科技有限公司 | A kind of infrared detection sensor regulating device |
| WO2019146582A1 (en) * | 2018-01-25 | 2019-08-01 | 国立研究開発法人産業技術総合研究所 | Image capture device, image capture system, and image capture method |
| CN110222310A (en) * | 2019-05-17 | 2019-09-10 | 科迈恩(北京)科技有限公司 | A kind of shared AI scientific instrument Data Analysis Services system and method |
| CN110490251A (en) * | 2019-03-08 | 2019-11-22 | 腾讯科技（深圳）有限公司 | Prediction disaggregated model acquisition methods and device, storage medium based on artificial intelligence |
| CN110793923A (en) * | 2019-10-31 | 2020-02-14 | 北京绿土科技有限公司 | Hyperspectral soil data acquisition and analysis method based on mobile phone |
| CN111609931A (en) * | 2020-05-17 | 2020-09-01 | 北京安洲科技有限公司 | Parameter-programmable real-time hyperspectral acquisition system and method |
-
2020
- 2020-10-13 CN CN202011088696.XA patent/CN112444493B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1995987A (en) * | 2007-02-08 | 2007-07-11 | 江苏大学 | Non-destructive detection method and device for agricultural and animal products based on hyperspectral image technology |
| WO2019146582A1 (en) * | 2018-01-25 | 2019-08-01 | 国立研究開発法人産業技術総合研究所 | Image capture device, image capture system, and image capture method |
| CN109341869A (en) * | 2018-12-04 | 2019-02-15 | 西安思科赛德电子科技有限公司 | A kind of infrared detection sensor regulating device |
| CN110490251A (en) * | 2019-03-08 | 2019-11-22 | 腾讯科技（深圳）有限公司 | Prediction disaggregated model acquisition methods and device, storage medium based on artificial intelligence |
| CN110222310A (en) * | 2019-05-17 | 2019-09-10 | 科迈恩(北京)科技有限公司 | A kind of shared AI scientific instrument Data Analysis Services system and method |
| CN110793923A (en) * | 2019-10-31 | 2020-02-14 | 北京绿土科技有限公司 | Hyperspectral soil data acquisition and analysis method based on mobile phone |
| CN111609931A (en) * | 2020-05-17 | 2020-09-01 | 北京安洲科技有限公司 | Parameter-programmable real-time hyperspectral acquisition system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112444493A (en) | 2021-03-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11542461B2 (en) | Analysis device | |
| EP2976605B1 (en) | Spectroscopic characterization of seafood | |
| CN104931470B (en) | A kind of pesticide residue detection device and detection method based on fluorescent high spectral technology | |
| CN104374738B (en) | A kind of method for qualitative analysis improving identification result based on near-infrared | |
| RU2388203C2 (en) | Device for detection of homogeneity in batch of seeds | |
| US11054370B2 (en) | Scanning devices for ascertaining attributes of tangible objects | |
| Eissa et al. | Understanding color image processing by machine vision for biological materials | |
| CN116416250B (en) | Finished product detecting system of fast food canned product production line | |
| US20220012515A1 (en) | System, method and computer program for guided image capturing of a meal | |
| CN112444493B (en) | Optical detection system and device based on artificial intelligence | |
| CN114663821A (en) | Real-time nondestructive detection method for product quality based on video hyperspectral imaging technology | |
| Srivastava et al. | Detection of insect damaged rice grains using visible and near infrared hyperspectral imaging technique | |
| US12007332B2 (en) | Portable scanning device for ascertaining attributes of sample materials | |
| CN114419311B (en) | Multi-source information-based passion fruit maturity nondestructive testing method and device | |
| CN114136920A (en) | Hyperspectrum-based single-grain hybrid rice seed variety identification method | |
| Xie et al. | Modeling for mung bean variety classification using visible and near-infrared hyperspectral imaging | |
| CN114295556A (en) | Hyperspectral physicochemical index fitting method and system based on one-dimensional convolution | |
| CN111077088B (en) | Smart phone imaging spectrometer and spectrum identification method thereof | |
| WO2019143316A1 (en) | Fluid classification | |
| CN107403181B (en) | Lean meat and fat meat self-adaptive separation method based on Guangdong style sausage hyperspectral image | |
| Groß et al. | Phas-maFOOD-A miniaturized multi-sensor solution for rapid, non-destructive food quality assessment | |
| CN114627120B (en) | Method for distinguishing decoloration of regenerated polyester material | |
| CN116308435B (en) | Agricultural product precise identification method and system based on intelligent traceability scale | |
| CN115015180B (en) | Fruit sugar degree nondestructive testing method based on spectral reflectance reconstruction technology | |
| Wang | Design of automatic mildew image detection system for corn grain |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |