CN116106235A

CN116106235A - Object identification method, device, equipment and system

Info

Publication number: CN116106235A
Application number: CN202111333862.2A
Authority: CN
Inventors: 马骞; 陆嘉文; 杨德良
Original assignee: Nanjing Weina Technology Research Institute Co ltd
Current assignee: Nanjing Kanshijie Intelligent Technology Co ltd
Priority date: 2021-11-11
Filing date: 2021-11-11
Publication date: 2023-05-12

Abstract

The invention provides an object identification method, device, equipment and system, wherein the method comprises the following steps: receiving an interference image of an object to be detected; determining spectrum information of an object to be detected according to the interference image, matching the spectrum information of the object to be detected with spectrum information of a matched object in a spectrum information base, and determining the matched object information which is successfully matched as the object information to be detected; the spectrum information base contains spectrum information of the paired objects, the obtained spectrum information is matched with the spectrum information of the paired objects in the spectrum information base based on the characteristic that the spectrum information of any object has uniqueness, so as to realize the identification of the object to be detected.

Description

Object identification method, device, equipment and system

Technical Field

The present invention relates to the field of image processing technologies, and in particular, to a method, an apparatus, a device, and a system for object identification.

Background

Object recognition is a fundamental study in the field of computer vision, whose task is to recognize what objects are in an image, and is widely used in a variety of scenes.

Generally, when identifying an object in an industrial field, one method is pattern matching, which is based on matching a two-dimensional image of the object with a template image, so as to obtain the type of the object to be detected; the other method is spectrum analysis, which is to obtain the class of the object to be measured by obtaining a hyperspectral image of the object to be measured and analyzing the curve of the reflectivity of the object to be measured along with the change of the wavelength according to the hyperspectral image.

However, when a two-dimensional image is subjected to pattern matching by a pattern matching method, recognition errors are easily caused by shape changes and the like; the method based on spectrum analysis needs a precise mechanical structure, and has the defects of high cost and difficult application and popularization in a large range. Therefore, there is a need to develop a new object recognition method to meet the requirements of low mechanical structure and high recognition accuracy.

Disclosure of Invention

The invention provides an object identification method, device, equipment and system, which are used for meeting the requirements of lower mechanical structure requirements and higher identification accuracy.

In a first aspect, an embodiment of the present invention provides an object recognition method, including:

receiving an interference image of an object to be detected;

determining spectral information of an object to be detected according to the interference image;

Matching the spectrum information of the object to be detected with the spectrum information of the matched object in the spectrum information base, and determining the matched object information successfully matched as the object information to be detected; the spectrum information base contains spectrum information of the paired objects.

Optionally, the interference image includes bright and dark fringes of different bright and dark degrees; or concentric rings of different shades.

Optionally, determining spectral information of the object to be measured according to the interference image includes:

when the interference image comprises light and dark fringes with different light and dark degrees, selecting the light and dark fringes with different light and dark degrees in the interference image;

and carrying out global information extraction on the selected bright and dark fringes, and restoring the extracted information according to the wavelength to obtain the transmittance under different wavelengths, wherein the obtained transmittance under different wavelengths is used as spectrum information.

Optionally, the object recognition method further includes:

and judging whether the paired object and the object to be detected are paired successfully or not.

Optionally, determining whether the pairing object and the object to be measured are successfully paired includes:

obtaining the similarity of the spectrum information of the object to be detected and the spectrum information of the matched object;

if the similarity is greater than or equal to a preset threshold, determining that the paired object and the object to be detected are paired successfully;

If the similarity is smaller than a preset threshold, determining that the pairing object and the object to be detected are failed to be paired.

Optionally, obtaining the similarity between the spectrum information of the object to be measured and the spectrum information of the counterpart object includes:

the similarity of the spectrum information of the object to be detected and the spectrum information of the matched object is determined by at least one of the following methods: neural network method, least square method and picture comparison method.

Optionally, the object recognition method further includes:

when all the paired objects in the spectrum information base are failed to be paired with the object to be detected, the user uploads the spectrum information of the object to be detected and the information of the object to be detected to the spectrum information base after determining the information of the object to be detected.

Optionally, the interference image is acquired and transmitted based on a spectral-spatial transformation device; the spectral-spatial conversion device includes: basic optical components for providing the required light for the filter cavity; a filter cavity for forming bright and dark fringes; an image sensor for forming an image; the surface of the filtering cavity is provided with a plurality of ladder structures.

Optionally, the filtering cavity is located between the base optical component and the image sensor; the position of the basic optical component in the spectrum space conversion device is one side close to the object to be detected.

Optionally, the size of the image sensor matches the size of the filter cavity.

Alternatively, the image sensor is a CMOS sensor or a CCD sensor.

Optionally, the filtering cavity comprises a transparent support; the upper surface and the lower surface of the transparent support piece are coated with reflective layer films; a step structure is arranged on the upper surface and/or the lower surface; the lower surface is the side opposite to the upper surface; the reflective layer film is used for: and reflecting a part of the light rays irradiated on the reflecting layer film.

Optionally, the upper surface is facing the base optical element and the lower surface is facing the image sensor.

Optionally, the transparent support comprises at least one of: glass, resin, transparent colloid;

the reflective layer film is made of at least one of the following materials: gold, silver, alumina and titania.

Optionally, the step structure includes steps, and the object recognition method further includes:

determining the step distribution on the surface of the filtering cavity according to the wave band range corresponding to the object to be detected and the level of the requirement on the image resolution of the interference image; the step profile includes the number and height of steps.

Optionally, the step distribution direction of the filtering cavity comprises a two-dimensional direction and a three-dimensional direction;

When the direction of the step distribution of the filtering cavity is a two-dimensional direction, a gradient exists in the first direction, and no gradient exists in the second direction;

when the direction of the step distribution of the filter cavity is a three-dimensional direction, then there is a first number of gradients in a first direction and a second number of gradients in a second direction.

Optionally, the object to be measured is a self-luminous object, or a light source is provided for the object to be measured through a light source device.

Optionally, the light source device comprises at least one of: LED light source, laser light source, halogen light source, solar simulation light source and fluorescent light source.

Optionally, the object to be tested is placed on a test rack, and the test rack is used for fixing the object to be tested.

Optionally, the interference image is an interference image obtained by the spectrum space conversion device when the spectrum space conversion device is placed right behind the object to be detected; wherein the right rear side is expressed as that when the light source device irradiates the object to be measured on one side of the object to be measured, the spectrum space conversion device is positioned on the other side opposite to the light source device;

or the interference image is obtained by the spectrum space conversion device when the spectrum space conversion device is placed at the side of the object to be detected; the side indicates that when the light source device irradiates the object to be measured on one side of the object to be measured, the spectrum space transforming device is located on the side of a straight line formed by the light source device and the object to be measured.

In a second aspect, an embodiment of the present invention provides an object recognition apparatus, including:

the receiving module is used for receiving the interference image of the object to be detected;

the determining module is used for determining spectral information of the object to be detected according to the interference image;

the matching module is used for matching the spectrum information of the object to be detected with the spectrum information of the matched object in the spectrum information base and determining the matched object information which is successfully matched as the object information to be detected; the spectrum information base contains spectrum information of the paired objects.

In a third aspect, an embodiment of the present invention provides an object recognition apparatus, including: at least one processor and memory;

the memory stores computer-executable instructions;

at least one processor executes computer-executable instructions stored in a memory, causing the at least one processor to perform the method as in any one of the first aspects.

In a fourth aspect, an embodiment of the present invention provides an object recognition system, including the object recognition apparatus and the spectral-spatial transformation device provided in the third aspect;

the spectrum space transformation device is used for acquiring interference images of the object to be detected and transmitting the interference images to the object identification equipment.

In a fifth aspect, an embodiment of the present invention provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements a method as in any of the first aspects.

In a sixth aspect, embodiments of the present invention provide a computer program product comprising a computer program which, when executed by a processor, implements a method as in any of the first aspects.

The embodiment of the invention provides an object identification method, device, equipment and system, wherein the method acquires an interference image of an object to be detected, further acquires spectrum information of the object to be detected, matches the acquired spectrum information of the object to be detected with spectrum information of a counterpart object in a spectrum information base based on the characteristic that the spectrum information of any object has uniqueness, so as to realize identification of the object to be detected.

Drawings

In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic diagram of an application scenario of object recognition according to an embodiment of the present invention;

fig. 2 is a schematic flow chart of an object recognition method according to an embodiment of the present invention;

FIG. 3 is a flowchart of another object recognition method according to an embodiment of the present invention;

fig. 4 is a schematic structural diagram of a spectrum space conversion device according to an embodiment of the present invention;

fig. 5a is a schematic structural diagram of a filtering cavity according to an embodiment of the present invention;

fig. 5b is a schematic structural diagram of a filtering cavity according to a second embodiment of the present invention;

fig. 5c is a schematic structural diagram III of a filtering cavity according to an embodiment of the present invention;

fig. 5d is a schematic structural diagram of a filtering cavity according to an embodiment of the present invention;

fig. 6 is a schematic structural diagram corresponding to a three-dimensional direction of the step distribution of the filtering cavity according to the embodiment of the present invention;

fig. 7 is a schematic structural diagram of an object recognition device according to an embodiment of the present invention;

fig. 8 is a schematic hardware structure of an object recognition device according to an embodiment of the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

The terms "first," "second," "third," "fourth" and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented, for example, in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

Fig. 1 is a schematic view of an application scenario of object recognition provided in an embodiment of the present invention, as shown in fig. 1, an interference image of an object to be detected is obtained through a spectrum space transformation device 101, and the obtained interference image is transmitted to an object recognition device 102, where the object recognition device 102 may process the interference image to obtain spectrum information of the object to be detected, determine information of the object to be detected based on the spectrum information, and output the determined information of the object to be detected. Wherein the interference image can be transmitted between the spectral space transformation apparatus 101 and the object recognition device 102 through a data transmission line. The object recognition device 102 may be an intelligent device such as a computer, a tablet, a computer, or other intelligent devices with data computing and processing functions.

In some technologies, a pattern matching or spectrum analysis method is generally adopted for object identification, but the method has the defects of lower identification accuracy or higher requirements on the mechanical structure of an instrument.

Based on the above-mentioned problems, the object recognition method provided by the embodiment of the present invention generates the interference image of the object to be detected by the device for generating the interference image, the object recognition device 102 may receive the interference image and acquire the corresponding spectrum information, and since the spectrum information of different objects has uniqueness, the information of the object to be detected may be determined based on the spectrum information, and the method does not recognize errors due to shape changes relative to the pattern matching method. For example, when the object to be detected is a piece of red paper, if the object to be detected is folded into a ship or an animal, the object to be detected can be accurately identified as the red paper by adopting the method of the invention; compared with a spectrum analysis method, the method can acquire the hyperspectral image without adopting a precise optical instrument, so that the method has the advantage of lower requirement on equipment on the premise of ensuring the accuracy of object identification; in addition, the identification method has higher identification efficiency.

The technical scheme of the invention is described in detail below by specific examples. The following embodiments may be combined with each other, and some embodiments may not be repeated for the same or similar concepts or processes.

Fig. 2 is a flowchart of an object recognition method according to an embodiment of the present invention, where the method is applied to an object recognition device 102, and the method includes:

step S201, receiving an interference image of the object to be detected.

The object recognition device 102 may receive an interference image of the object under test from a device that generates the interference image, wherein the device that generates the interference image may generate the interference image of the object under test. The device for generating an interference image is provided with a device for forming an interference image so that when light is irradiated, an interference image can be obtained. For example, the interference image includes stripe structures of different shades, or the interference image includes non-stripe structures of different shades. Wherein the thickness of the stripes is randomly generated. Wherein the device generating the interference image can obtain the interference image according to the interference principle of light.

When the interference image is in a non-fringe structure, the interference image may be in a concentric ring structure. The specific structure of the interference image may be determined based on the structure of the filter array points (filter cavities 2) in the device that generated the interference image. When the structure of the filtering cavity 2 includes a planar glass and a plano-convex lens, and the convex surface of the plano-convex lens is contacted with the planar glass, an interference image of a circular ring with alternate brightness and darkness can be obtained under the irradiation of the light source, wherein the contact point is a dark point.

Step S202, determining spectral information of the object to be detected according to the interference image.

After receiving the interference image, the interference image may be processed to obtain spectral information of the object under certain optical environment.

Specifically, when the spectrum information is acquired, global information in the interference image, such as gray value information of pixels, can be extracted to obtain the spectrum information of the object to be detected.

Step S203, matching the spectrum information of the object to be detected with the spectrum information of the matched object in the spectrum information base, and determining the matched object information which is successfully matched as the object information to be detected; the spectrum information base contains spectrum information of the paired objects.

The object is identified based on the spectrum information, wherein the object is identified based on that different substances can form a specific spectrum under a certain optical environment, and different objects can be identified according to the formed spectrum.

Specifically, a spectrum information base may be generated in advance, and spectrum information of the counterpart object is stored in the spectrum information base. After the spectrum information of the object to be detected is obtained, comparing the spectrum information of the object to be detected with the spectrum information of all the paired objects in the spectrum information base, and screening the spectrum information of the paired objects matched with the spectrum information of the object to be detected in the spectrum information base.

And after screening out the spectrum information of the matched object matched with the spectrum information of the object to be tested, determining the screened matched object information as the object information to be tested.

For example, when the object to be measured is a red card paper, the spectrum information obtained after the processing of the device for generating interference images and the object recognition device 102 is generated by the object to be measured in the illumination environment of 380nm-780nm wave band. The spectrum information base stores various spectrum information of the object under the wave band, and if the spectrum information of the object to be detected is matched with the spectrum information of the object A in the spectrum information base through comparison, and the object A is red card paper, the object to be detected can be determined to be red card paper.

According to the method, the interference image of the object to be detected is received, the spectrum information of the object to be detected is determined according to the interference image, the spectrum information of the object to be detected is matched with the spectrum information of the matched object in the spectrum information base, and the matched object information which is successfully matched is determined as the object information to be detected, so that the object to be detected can be identified based on the spectrum information. In addition, the identification method has higher identification efficiency.

Fig. 3 is a flow chart of another object recognition method according to an embodiment of the present invention, as shown in fig. 3, the method includes:

s301, receiving an interference image of an object to be detected.

S302, selecting light and shade fringes with different light and shade degrees in the interference image, wherein the interference image comprises the light and shade fringes with different light and shade degrees.

When the interference image includes a plurality of bright-dark fringes, the bright-dark fringes in the interference image may be selected after receiving the interference image, because spectral information exists in the bright-dark fringes, information may be extracted from the selected bright-dark fringes. In order to facilitate the extraction of information, the bright and dark stripes need to be selected first, and then only the global information of the selected part can be extracted in subsequent processing.

S303, carrying out global information extraction on the selected bright and dark fringes, reducing the extracted information according to the wavelength to obtain the transmittance under different wavelengths, and taking the obtained transmittance under different wavelengths as spectrum information.

Wherein the interference image comprises bright and dark fringes with different brightness degrees.

After the bright and dark stripes are selected, global information can be extracted, namely gray value information of pixels in a selected area is obtained, the obtained gray values of the pixels are restored according to wavelengths, the transmittance under different wavelengths is obtained, and finally spectrum information is obtained.

S304, the spectrum information of the object to be detected is matched with the spectrum information of the matched object in the spectrum information base.

S305, judging whether the paired object and the object to be detected are paired successfully.

When judging whether the pairing object and the object to be measured are successfully paired, the following implementation mode can be adopted:

optionally, obtaining the similarity between the spectrum information of the object to be detected and the spectrum information of the matched object; if the similarity is greater than or equal to a preset threshold, determining that the paired object and the object to be detected are paired successfully; if the similarity is smaller than a preset threshold, determining that the pairing object and the object to be detected are failed to be paired.

When judging whether pairing is successful, acquiring the similarity between the spectrum information of the object to be detected and the spectrum information of all paired objects in the spectrum information library, acquiring a highest similarity value, comparing the highest similarity value with a preset threshold value, and if the highest similarity value is greater than or equal to the preset threshold value, determining that pairing is successful, wherein the paired object information corresponding to the spectrum information which is successfully paired is the object information to be detected.

If the highest similarity value is smaller than the preset threshold value, the matching failure is indicated, namely, the spectrum information of the matched object matched with the spectrum information of the object to be detected does not exist in the spectrum information base.

For example, when the determined highest similarity value is 98%, if the preset threshold value is 95%, the matching may be determined to be successful, and if the determined highest similarity value is 80%, the matching may be determined to be failed.

By setting the threshold value, whether spectrum information of a matched object matched with spectrum information of an object to be detected exists in the spectrum information base can be accurately determined.

Optionally, obtaining the similarity between the spectrum information of the object to be measured and the spectrum information of the counterpart object includes: the similarity of the spectrum information of the object to be detected and the spectrum information of the matched object is determined by at least one of the following methods: neural network method, least square method and picture comparison method.

Wherein, the spectrum information is: the abscissa is wavelength, and the ordinate is transmittance, that is, the transmittance of the object to be measured at different wavelengths, may form a curve. When the similarity of the spectrum information of the object to be measured and the spectrum information of the counterpart object is obtained, the similarity of the two curves is substantially obtained.

For the problem of obtaining the similarity of the two curves, a neural network method can be adopted, for example, the characteristic values of the two curves are obtained through the neural network, and the similarity is determined by comparing the characteristic values. And the distance between the two curves can be obtained by adopting a least square method, so that the similarity of the two curves can be measured according to the distance. In addition, a picture comprising two curves and an abscissa and an ordinate can be generated, and the similarity of the two pictures can be directly compared.

The method can accurately obtain the similarity between the two spectrum information, and can accurately identify the object to be detected based on the determined similarity.

S306, determining the paired object information successfully paired as the object information to be detected.

When the pairing is successful, the paired object information of the spectrum information base, which is successfully paired, can be obtained, and the paired object information is determined to be the object information to be detected. For example, when it is determined that the paired object information for which pairing is successful is a red jam, it may be determined that the object to be measured is a red jam.

S307, when all the paired objects in the spectrum information base are failed to be paired with the object to be detected, the user uploads the spectrum information of the object to be detected and the information of the object to be detected to the spectrum information base after determining the information of the object to be detected.

When all the paired objects in the spectrum information base are failed to be paired with the object to be detected, namely the spectrum information of the object to be detected does not exist in the spectrum information base, at the moment, the information of the object to be detected can be determined in other modes. For example, methods based on raman spectroscopy identify different objects to be measured. After the information of the object to be detected is determined, the acquired spectrum information of the object to be detected and the information of the object to be detected can be uploaded to a spectrum information base, so that the spectrum information stored in the spectrum information base is more abundant in variety, and the information of the object to be detected can be accurately identified when the object to be detected is identified later.

The spectrum information in the spectrum information base can be enriched by uploading the information of the object to be detected and the spectrum information of the object to be detected which are failed to be paired to the spectrum information base, so that the types of identifiable objects to be detected are increased.

Two specific object recognition processes are described below.

If the object to be measured is paper with various colors, the paper to be measured can be placed on a test frame, when the light source device such as an LED light source or a laser light source is adopted to irradiate the paper, after the interference image is obtained through the spectrum space conversion device 101, the interference image is transmitted to the object recognition device 102, such as a computer, the object recognition device 102 performs global information extraction on the interference image to obtain spectrum information of the object to be measured, and compares the spectrum information of the object to be measured with spectrum information of a counterpart object stored in a spectrum information base to obtain similarity, and when the similarity is greater than or equal to a preset threshold, if the counterpart object information corresponding to the similarity is red paper, the object to be measured can be determined to be red paper, wherein the spectrum information of the paper with different colors is stored in the spectrum information base.

If the object to be detected is a different light source, the light source to be identified can be placed on the test frame, the light source to be identified can be an LED light source, a fluorescent light source, a halogen light source, a laser light source and the like, and the specific processing process is similar to the above process, wherein the spectrum information of the different types of light sources is stored in the spectrum information base.

In addition, the spectrum information obtained by the same object under different light sources is different. Therefore, before the object identification, the spectrum information of the matched object stored in the spectrum information base is obtained under the irradiation of the light source A, and the spectrum information of the object to be detected is also obtained under the irradiation of the light source A, so that after the spectrum information of the object to be detected is obtained, the spectrum information can be directly compared with the spectrum information of the matched object in the spectrum information base, and the object to be detected is identified.

Or when the spectrum information of the object to be measured is obtained under the irradiation of the light source A, and the spectrum information of the matched object stored in the spectrum information base is obtained under the irradiation of the light source B, a normalization coefficient can be firstly obtained, the normalization coefficient is equal to the result of dividing the spectrum information of the light source A and the spectrum information of the light source B, and after the spectrum information 1 of the object to be measured under the light source A is obtained, the spectrum information 1 is multiplied by the normalization coefficient and converted into the spectrum information 2 of the object to be measured under the light source B. And comparing the converted spectrum information 2 with spectrum information of each matched object stored in a spectrum information base to identify the object to be detected.

In one embodiment, the interference image is acquired and transmitted based on the spectral-spatial transformation device 101. The structure of the spectral-spatial conversion device 101 that acquires and transmits an interference image is described in detail below.

Fig. 4 is a schematic structural diagram of a spectrum space conversion device 101 according to an embodiment of the present invention, and as shown in fig. 4, the spectrum space conversion device 101 includes: a base optical component 1 for providing the required light for the filter cavity 2; a filter cavity 2 for forming bright and dark fringes; an image sensor 3 for forming an image; the surface of the filter cavity 2 is provided with a plurality of ladder-like structures.

The basic optical component 1 may be a lens, which can well achieve that light is uniformly irradiated on the surface of the filtering cavity 2. In addition, in order to reduce the reflection of the lens surface, an antireflection film can be plated on the surface of the lens, so that imaging is clearer.

The filter Cavity 2 is essentially a F-P Cavity (Fabry-perot Cavity) and may produce bright and dark fringes. The side surface of the filtering cavity 2 can be provided with a plurality of ladder structures, so that when the light beam irradiates the side surface of the filtering cavity 2, transmitted light and reflected light can be generated, the transmitted light is reflected and refracted again when passing through the reflective layer film on the other surface of the filtering cavity 2, and the reflected light of the two times is light with the same frequency, so that interference phenomenon can be generated. When the number and the height of the stepped structures arranged on the surface of the filtering cavity 2 are different, the image resolution of the interference image and whether the interference image in the wave band range corresponding to the object to be detected can be obtained can be influenced. It should be noted that, a plurality of ladder structures may be disposed on the other surface of the filter cavity 2.

In addition, the spectrum space conversion device 101 further includes an image sensor 3, and the image sensor 3 may receive an optical signal passing through an object to be measured and convert the optical signal into an interference image, and specifically, the image sensor 3 may be controlled in a software control manner, so that the image sensor 3 may save a bright-dark stripe currently formed in the form of an image, for example, by setting a memory card to save the bright-dark stripe in the form of an image. Also, the saved interference image may be transmitted to the object recognition device 102 shown in fig. 1 so that the object recognition device 102 performs processing based on the interference image.

Optionally, the filter cavity 2 is located between the base optical component 1 and the image sensor 3; the position of the basic optical element 1 in the spectral space transformation apparatus 101 is a side close to the object to be measured.

Wherein the positions of the individual components in the spectral-spatial transformation apparatus 101 also need to be defined. The basic optical component 1 needs to be arranged on one side close to the object to be detected, so that the light rays emitted or reflected by the object to be detected are processed firstly, the light rays required by the filtering cavity 2 are obtained, the filtering cavity 2 is placed behind the basic optical component 1, the processed light rays emitted or reflected by the object to be detected can be generated into bright and dark stripes, and then an interference image is generated through the image sensor 3.

Optionally, the size of the image sensor 3 is matched to the size of the filter cavity 2.

In addition, in order to enable the image sensor 3 to image the bright and dark fringes formed by the filtering cavity 2, the size of the image sensor 3 can be matched with the size of the filtering cavity 2, so that the situation that a complete interference image cannot be acquired is avoided. Wherein, the matching means: in all directions, the photosensitive size of the image sensor 3 is equal to or larger than the effective size of the filter cavity 2. The effective size refers to a size corresponding to a part of the filtering cavity 2 provided with the ladder-type structure.

Alternatively, the image sensor 3 is a CMOS sensor or a CCD sensor.

The CMOS sensor and the CCD sensor are commonly used digital image sensors 3, and the CMOS sensor and the CCD sensor have differences in cost, sensitivity, power consumption ratio, transmission speed, and the like, so that the image sensor 3 can be selected according to the need of identifying an object to be detected.

Optionally, the filtering cavity 2 comprises a transparent support; the upper surface and the lower surface of the transparent support piece are coated with reflective layer films; a step structure is arranged on the upper surface and/or the lower surface; the lower surface is the side opposite to the upper surface; the reflective layer film is used for: and reflecting a part of the light rays irradiated on the reflecting layer film.

For the filtering cavity 2, the filtering cavity 2 is composed of a transparent support and a reflective layer film, the transparent support comprises an upper surface and a lower surface, and the shape of the upper surface of the transparent support can be set into a ladder type structure according to requirements. Wherein, in order to realize the reflection of light, a reflective layer film needs to be coated on the upper surface of the transparent support. For the lower surface, a reflective layer film is also required to be coated in order to achieve reflection of light.

The above-described manner of providing the stepped structure on the upper surface is merely an example, and in practice, the stepped structure may be provided on the lower surface. When only set up the ladder type structure at the upper surface, the one side can have the difference in height of two directions, when also setting up the ladder type structure at the lower surface, then can have the difference in height of four directions at two sides, and its technology that realizes is more complicated, and the advantage is that the image resolution ratio that the spectral information that acquires corresponds is higher.

Optionally, the upper surface faces the base optical component 1 and the lower surface faces the image sensor 3.

Since the filter cavity 2 is divided into an upper surface and a lower surface, when the filter cavity 2 is disposed between the base optical component 1 and the image sensor 3, the upper surface of the filter cavity 2 needs to be close to the base optical component 1 and the lower surface of the filter cavity 2 needs to be close to the image sensor 3.

In some embodiments, the transparent support is made of a material that has a certain requirement, and the transparent support needs to have a good penetrability for light. Therefore, the transparent support member can be made of glass, resin, transparent colloid, etc., and the light transmittance of the material is high, so that the attenuation of the optical signal can be reduced. When the material is resin and transparent colloid, the hardness of the resin and the hardness of the transparent colloid are required to meet certain requirements.

In addition, the material to the reflection of light layer membrane also has certain requirement, and when the material of reflection of light layer membrane is: when one of gold, silver, aluminum oxide and titanium dioxide is used, the light has better reflection effect.

determining the step distribution on the surface of the filter cavity 2 according to the wave band range corresponding to the object to be detected and the level of the requirement on the image resolution of the interference image; the step profile includes the number and height of steps.

In some embodiments, interference images of different image resolutions and different band ranges may be obtained for different step distributions, i.e., different numbers and heights of steps. For example, fig. 5 (a) to 5 (d) are schematic structural diagrams of four different types of filter cavities 2 according to an embodiment of the present invention. Wherein, the resolution of the image corresponding to the filtering cavity 2 in fig. 5 (a) is less than 1nm; the image resolution corresponding to the filter cavity 2 of fig. 5 (b) is less than 0.6nm; the image resolution corresponding to the filter cavity 2 of fig. 5 (c) is less than 10nm; the image resolution corresponding to the filter cavity 2 of fig. 5 (d) is less than 30nm. Wherein the image resolution is higher as the difference in height of the steps is smaller. The number and the height of the steps on the surface of the filtering cavity 2 can influence the wave band range and the image resolution corresponding to the identifiable object to be detected.

For example, when the object to be measured has reflection peaks at 550nm, 550.5nm and 551nm, it is necessary to employ the filter cavity 2 capable of recognizing the reflection peak at 550.5nm with a wavelength of less than 1 nm.

Because the wave band ranges corresponding to different objects to be detected are different, and the requirements on the image resolution are different when the object identification is carried out, the results are related to the number and the height of the steps on the surface of the filtering cavity 2, and the number and the height of the steps distributed on the surface of the filtering cavity 2 can be determined according to the requirements on the image resolution of the interference image and the wave band range corresponding to the objects to be detected.

Optionally, the direction of the step distribution of the filtering cavity 2 includes a two-dimensional direction and a three-dimensional direction;

when the direction of the step distribution of the filter cavity 2 is a two-dimensional direction, a gradient exists in the first direction, and no gradient exists in the second direction;

when the direction of the step distribution of the filter cavity 2 is a three-dimensional direction, then there is a first number of gradients in a first direction and a second number of gradients in a second direction.

The step distribution of the filter cavity 2 may be two-dimensional, and there is a gradient in the first direction, and there is no gradient in the second direction, as shown in the schematic structural diagrams of the four filter cavities 2 shown in fig. 5 (a) to 5 (d).

Fig. 6 is a schematic structural diagram corresponding to a three-dimensional direction of the step distribution of the filtering cavity 2 according to the embodiment of the present invention, where, as shown in fig. 6, the first direction is a direction, the second direction is b direction, gradients exist in the a direction and the b direction at the same time, and when 10 gradients exist in the a direction and 10 gradients exist in the b direction, 10×10=100 gradient distributions can be realized in the three-dimensional direction.

By setting the direction of the step distribution of the filter cavity 2 to the three-dimensional direction, the degree of density of the gradient distribution, that is, the number and height of steps can be increased to improve the image resolution of the generated interference image.

In some embodiments, the object to be measured may be a self-luminous object, so that the spectral-spatial transformation device 101 may acquire an interference image of the object to be measured; alternatively, a light source may be provided for the object to be measured, so that the spectral-spatial transformation device 101 may acquire an interference image of the object to be measured under irradiation of the light source.

In order to acquire clear interference images, an object to be measured can be placed on the test frame, so that the phenomenon that the object to be measured is in a shaking state when the object to be measured is an irregularly-shaped object is avoided, and the acquired interference images are not clear is avoided.

Alternatively, when the spectrum space conversion device 101 is placed right behind the object to be measured, the spectrum space conversion device 101 obtains the interference image; wherein the right rear is represented as when the light source device irradiates the object to be measured on one side of the object to be measured, the spectrum space transforming device 101 is located on the other side opposite to the light source device;

alternatively, the interference image is an interference image obtained by the spectrum space conversion device 101 when the spectrum space conversion device 101 is placed on the side of the object to be measured; the side indicates that when the light source device irradiates the object to be measured on one side of the object to be measured, the spectral-spatial conversion device 101 is located on the side of a straight line formed by the light source device and the object to be measured.

In some techniques, one of the obtained interference images is an interference image obtained after light is transmitted through an object to be measured, and the other is an interference image obtained after light is irradiated to the surface of the object to be measured and then subjected to diffuse reflection. The above methods for acquiring interference images are not performed simultaneously, and one method for acquiring interference images is selected according to needs. Wherein for transmitted light, the information of the object to be detected can be determined based on the corresponding interference image, for example, whether the interior of the apple is deteriorated or not is identified. For diffusely reflected light, information of the object surface may be determined based on the corresponding interference image, e.g. the presence of black spots on the apple surface may be identified.

When the spectrum space conversion device 101 is placed at the side of the object to be measured, one implementation manner is as follows: the light source device irradiates the object to be measured vertically, and the angle between the line between the spectral space transformation apparatus 101 and the object to be measured and the line between the light source device and the object to be measured is 30 degrees to 60 degrees, preferably, the angle may be 45 degrees.

According to the detected difference of the object to be detected, a corresponding interference image can be obtained, and the purpose of accurately determining the information of the object to be detected is achieved.

According to the method, the interference image of the object to be detected is received, the spectrum information of the object to be detected is determined according to the interference image, the spectrum information of the object to be detected is matched with the spectrum information of the matched object in the spectrum information base, and the matched object information is determined to be the object information to be detected, so that the object to be detected can be identified based on the spectrum information. On the basis, the structure of the optical space transformation device 101 is improved, such as the matching process of the filtering cavity 2 and the image sensor 3 or between the filtering cavity and the image sensor 3 is improved, so that the acquired interference image is clearer, the accuracy of the acquired optical spectrum information is improved, and the accuracy of object identification is further improved.

Fig. 7 is a schematic structural diagram of an object recognition device according to an embodiment of the present invention; as shown in fig. 7, the object recognition apparatus includes:

a receiving module 701, configured to receive an interference image of an object to be detected;

a determining module 702, configured to determine spectral information of an object to be detected according to the interference image;

the matching module 703 is configured to match spectrum information of an object to be detected with spectrum information of a counterpart object in the spectrum information base, and determine paired object information that is successfully paired as the object information to be detected; the spectrum information base contains spectrum information of the paired objects.

Optionally, the interference image includes bright and dark fringes of different bright and dark degrees; the determining module 702 is specifically configured to:

selecting light and shade stripes with different light and shade degrees in the interference image;

and carrying out global information extraction on the bright and dark fringes, reducing the extracted information according to the wavelength to obtain the transmittance under different wavelengths, and taking the obtained transmittance under different wavelengths as spectrum information.

Optionally, the object recognition device further includes: the judging module is used for:

Optionally, the judging module is specifically configured to:

Optionally, when the judging module obtains the similarity between the spectrum information of the object to be detected and the spectrum information of the counterpart object, the judging module is specifically configured to:

Optionally, the apparatus further includes an upload module, configured to:

Alternatively, the interference image is acquired and transmitted based on the spectral-spatial transformation means 101; the spectral-spatial conversion device 101 includes: a base optical component 1 for providing the required light for the filter cavity 2; a filter cavity 2 for forming bright and dark fringes; an image sensor 3 for forming an image; the surface of the filter cavity 2 is provided with a plurality of ladder-like structures.

Alternatively, the image sensor 3 is a CMOS sensor or a CCD sensor.

determining the distribution of steps on the surface of the filtering cavity according to the wave band range corresponding to the object to be detected and the level of the requirement on the image resolution of the interference image; the distribution of steps includes: the number and height of the steps.

Alternatively, the interference image is an interference image obtained by the spectrum space conversion device 101 when the spectrum space conversion device 101 is placed right behind the object to be measured; wherein the right rear is represented as when the light source device irradiates the object to be measured on one side of the object to be measured, the spectrum space transforming device 101 is located on the other side opposite to the light source device;

The object recognition device provided by the embodiment of the invention corresponds to the object recognition method in the above embodiment, and its implementation principle and technical effects are similar, and will not be described here again.

Fig. 8 is a schematic hardware structure of an object recognition device 80 according to an embodiment of the present invention, as shown in fig. 8, where the object recognition device 80 provided in this embodiment includes: at least one processor 801 and a memory 802. The processor 801 and the memory 802 are connected by a bus 803.

In a specific implementation, the at least one processor 801 executes computer-executable instructions stored in the memory 802, so that the at least one processor 801 performs the object recognition method in the above-described method embodiment.

The specific implementation process of the processor 801 may refer to the above-mentioned method embodiment, and its implementation principle and technical effects are similar, and this embodiment will not be described herein again.

In the embodiment shown in fig. 8, it should be understood that the processor 801 may be a central processing unit (in english: central Processing Unit, abbreviated as CPU), or may be other general purpose processors, digital signal processors (in english: digital Signal Processor, abbreviated as DSP), application specific integrated circuits (in english: application Specific Integrated Circuit, abbreviated as ASIC), or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the present invention may be embodied directly in a hardware processor for execution, or in a combination of hardware and software modules in a processor for execution.

Memory 802 may comprise high-speed RAM memory or may also include nonvolatile storage NVM, such as at least one disk memory.

The bus 803 may be an industry standard architecture (Industry Standard Architecture, ISA) bus, an external device interconnect (Peripheral Component, PCI) bus, or an extended industry standard architecture (Extended Industry Standard Architecture, EISA) bus, among others. The buses may be divided into address buses, data buses, control buses, etc. For ease of illustration, the buses in the drawings of the present application are not limited to only one bus or one type of bus.

The embodiment of the present invention further provides a computer readable storage medium, in which computer executable instructions are stored, and when the processor 801 executes the computer executable instructions, the object recognition method of the above method embodiment is implemented.

The computer readable storage medium described above may be implemented by any type or combination of volatile or nonvolatile memory devices such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disk. A readable storage medium can be any available medium that can be accessed by a general purpose or special purpose computer.

An exemplary readable storage medium is coupled to the processor such the processor 801 can read information from, and write information to, the readable storage medium. In the alternative, the readable storage medium may be integral to the processor. The processor 801 and the readable storage medium may be located in an application specific integrated circuit (Application Specific Integrated Circuits, abbreviated as ASIC). Of course, the processor 801 and the readable storage medium may also be present as discrete components in a device.

An embodiment of the present application provides a computer program product comprising a computer program which, when executed by a processor, implements the object recognition method as provided by the embodiments corresponding to fig. 2 and 3 of the present application.

Those of ordinary skill in the art will appreciate that: all or part of the steps for implementing the method embodiments described above may be performed by hardware associated with program instructions. The foregoing program may be stored in a computer readable storage medium. The program, when executed, performs steps including the method embodiments described above; and the aforementioned storage medium includes: various media that can store program code, such as ROM, RAM, magnetic or optical disks.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application.

Claims

1. An object recognition method, comprising:

receiving an interference image of an object to be detected;

determining spectral information of the object to be detected according to the interference image;

matching the spectrum information of the object to be detected with the spectrum information of the matched object in the spectrum information base, and determining the matched object information successfully matched as the object information to be detected; the spectrum information base comprises spectrum information of the matched object.

2. The method of claim 1, wherein the interference image comprises bright-dark fringes of different degrees of darkness; or concentric rings of different shades.

3. The method of claim 2, wherein the step of determining the position of the substrate comprises,

determining spectral information of the object to be detected according to the interference image, including:

and carrying out global information extraction on the selected bright and dark fringes, restoring the extracted information according to the wavelength to obtain the transmittance under different wavelengths, and taking the obtained transmittance under different wavelengths as spectrum information.

4. The method according to claim 1, wherein the method further comprises:

5. The method of claim 4, wherein determining whether the pairing of the paired object and the object under test is successful comprises:

if the similarity is greater than or equal to a preset threshold, determining that the pairing object and the object to be detected are successfully paired;

and if the similarity is smaller than the preset threshold, determining that the pairing of the paired object and the object to be detected fails.

6. The method of claim 5, wherein obtaining the similarity of the spectral information of the object to be measured and the spectral information of the counterpart object comprises:

determining the similarity of the spectrum information of the object to be detected and the spectrum information of the matched object by at least one of the following methods: neural network method, least square method and picture comparison method.

7. The method of claim 5, wherein the method further comprises:

when all the paired objects in the spectrum information base are failed to be paired with the object to be detected, after determining the information of the object to be detected, a user uploads the spectrum information of the object to be detected and the information of the object to be detected to the spectrum information base.

8. A method according to claim 3, wherein the interference image is acquired and transmitted based on a spectral-spatial transformation means; the spectral-spatial transformation apparatus includes: basic optical components for providing the required light for the filter cavity; a filter cavity for forming bright and dark fringes; an image sensor for forming an image; the surface of the filtering cavity is provided with a plurality of ladder structures.

9. The method of claim 8, wherein the filter cavity is located between the base optical component and the image sensor; the position of the basic optical component in the spectrum space conversion device is one side close to the object to be detected.

10. The method of claim 8, wherein the size of the image sensor matches the size of the filter cavity.

11. The method of claim 8, wherein the image sensor is a CMOS sensor or a CCD sensor.

12. The method of any one of claims 8-11, wherein the filter cavity comprises a transparent support; the upper surface and the lower surface of the transparent support piece are coated with reflective layer films; a step structure is arranged on the upper surface and/or the lower surface; the lower surface is the side opposite to the upper surface; the reflecting layer film is used for: and reflecting a part of light rays irradiated on the reflecting layer film.

13. The method of claim 12, wherein the upper surface is facing the base optical element and the lower surface is facing the image sensor.

14. The method of claim 12, wherein the transparent support comprises at least one of: glass, resin, transparent colloid;

15. The method of any one of claims 8-11, wherein the stair-step type structure comprises a stair, the method further comprising:

determining step distribution on the surface of the filtering cavity according to the wave band range corresponding to the object to be detected and the level of the requirement on the image resolution of the interference image; the step distribution includes: the number and height of the steps.

16. The method of claim 15, wherein the direction of the step profile of the filter cavity comprises a two-dimensional direction and a three-dimensional direction;

when the step distribution direction of the filtering cavity is a two-dimensional direction, a gradient exists in the first direction, and no gradient exists in the second direction;

when the direction of the step distribution of the filtering cavity is a three-dimensional direction, a first number of gradients exist in a first direction, and a second number of gradients exist in a second direction.

17. The method according to claim 8, wherein the object to be measured is a self-luminous object or a light source is provided to the object to be measured by a light source device.

18. The method of claim 17, wherein the light source device comprises at least one of: LED light source, laser light source, halogen light source, solar simulation light source and fluorescent light source.

19. The method of claim 17, wherein the object to be tested is placed on a test rack, the test rack being used to hold the object to be tested.

20. The method according to claim 17, wherein the interference image is an interference image obtained by the spectral-spatial transformation device when the spectral-spatial transformation device is placed directly behind the object to be measured; wherein the right rear side is expressed as that when the light source device irradiates the object to be measured on one side of the object to be measured, the spectrum space transforming device is positioned on the other side opposite to the light source device;

or the interference image is obtained by the spectrum space conversion device when the spectrum space conversion device is placed at the side of the object to be detected; the side means that when the light source device irradiates the object to be measured on one side of the object to be measured, the spectrum space transforming device is located on the side of a straight line formed by the light source device and the object to be measured.

21. An object recognition device, the device comprising:

the determining module is used for determining the spectrum information of the object to be detected according to the interference image;

the matching module is used for matching the spectrum information of the object to be detected with the spectrum information of the matched object in the spectrum information base and determining the matched object information which is successfully matched as the object information to be detected; the spectrum information base comprises spectrum information of the matched object.

22. An object recognition apparatus, characterized by comprising: at least one processor and memory;

the memory stores computer-executable instructions;

the at least one processor executing computer-executable instructions stored in the memory causes the at least one processor to perform the method of any one of claims 1-20.

23. An object recognition system comprising the object recognition apparatus of claim 22 and a spectral-spatial transformation device;

the spectrum space conversion device is used for acquiring interference images of the object to be detected and transmitting the interference images to the object identification equipment.

24. A computer readable storage medium, characterized in that the storage medium stores a computer program which, when executed by a processor, implements the method according to any of claims 1-20.

25. A computer program product comprising a computer program which, when executed by a processor, implements the method of any of claims 1-20.