RU2708454C1 - Method of forming, transmitting and restoring signals of different-spectral images - Google Patents

Method of forming, transmitting and restoring signals of different-spectral images Download PDF

Info

Publication number
RU2708454C1
RU2708454C1 RU2018137166A RU2018137166A RU2708454C1 RU 2708454 C1 RU2708454 C1 RU 2708454C1 RU 2018137166 A RU2018137166 A RU 2018137166A RU 2018137166 A RU2018137166 A RU 2018137166A RU 2708454 C1 RU2708454 C1 RU 2708454C1
Authority
RU
Russia
Prior art keywords
λ
signals
spectral
images
different
Prior art date
Application number
RU2018137166A
Other languages
Russian (ru)
Inventor
Юрий Сагдуллаевич Сагдуллаев
Сергей Дмитриевич Ковин
Константин Григорьевич Жуковский
Original Assignee
Закрытое акционерное общество "МНИТИ" (ЗАО "МНИТИ")
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Закрытое акционерное общество "МНИТИ" (ЗАО "МНИТИ") filed Critical Закрытое акционерное общество "МНИТИ" (ЗАО "МНИТИ")
Priority to RU2018137166A priority Critical patent/RU2708454C1/en
Application granted granted Critical
Publication of RU2708454C1 publication Critical patent/RU2708454C1/en

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00Television systems
    • H04N7/18Closed circuit television systems, i.e. systems in which the signal is not broadcast

Abstract

FIELD: electrical communication engineering.
SUBSTANCE: invention relates to spectrozonal television using registration of reflected or emitted flux in several zones of optical spectrum. Result is achieved by the fact that after splitting of input radiant flux F(λ) into two identical streams and passing them through two broadband optical filters ΟF1 and OF2, spectral characteristic of which is mutually opposite and satisfies the condition F1(λ)=1-F2(λ) in a spectral region with a wavelength from λ1 up to λn, and converting radiant fluxes using two matrix photodetectors having the same rectangular spectral characteristic in a spectral region with a wavelength from λ1 up to λn, generating two initial integral image signals U1(t) and U2(t), further separately for first and second integral image signals U1(t) and U2(t) based on analysis of their current amplitude values is generated for time moment Δti, corresponding to one i image element, m signals of different-spectral images by sampling from a pre-prepared and stored array of data in a memory device, which include a plurality of amplitude signal values for each detection region Δλi, then obtained signals of different-spectrum images are processed, selecting from any m signals of different spectral images any three signals of different-spectral images, sending said signals to inputs of a color video monitoring device and displaying said signals for visual analysis, as well as automatic selection of specified objects based on analysis of distribution of amplitude values m signals of different-spectral images within spectral section with wavelength from λ1 up to λn.
EFFECT: providing the possibility of generating signals of different-spectrum images for several zones of detecting a radiant flux inside a wide spectral region and increasing reliability of detection, selection and recognition of objects.
1 cl, 2 tbl, 1 dwg

Description

The present invention relates to the field of spectrozonal television, using the registration of reflected or radiated radiant flux in several zones of the ultraviolet (UV), visible (VI) or infrared (IR) regions of the spectrum, and can be used to solve problems of detection, selection and recognition of objects by their spectral and energetic features and find application in remote sensing systems for objects to automatically monitor the state or change the spectral characteristics of objects in space and others.

The issues of spectral selection of optical images were reflected in [1-3], which discusses the principles of constructing spectrozonal television (TV) systems for remote sensing of the Earth and shows the selection of radiant flux detection zones for selection and classification of objects with their known spectral characteristics and reflects the main questions of spectral selection and recognition of objects. The principle of spectral selection of optical objects using television is based on recording the reflected or radiated radiant flux in several spectral zones Δλ i located inside a wide spectral region from λ 1 to λ n . Moreover, depending on the problem being solved, the number of registration zones can be equal to m = 2, 3, 4, ..., R.

As you know, depending on the type of objects, the magnitude of the reflected or radiated radiant flux will be unequal in a given spectral region from λ 1 to λ n . Therefore, the choice of the number of zones of registration of the radiant flux depending on the number and classes of objects in the observed space plays an important role in solving problems of spectral selection [2]. Each object of the material world (natural or artificial) is characterized by its spectral reflection coefficient of the incident radiant (light) flux in the spectral region of wavelengths from λ 1 to λ n , which is equal to

ρ (λ) = F o (λ) / F p (λ),

where F p (λ) is the incident radiant (light) flux from the radiation source, F o (λ) is the reflected radiant (light) flux from objects.

The coefficient ρ (λ) characterizes the magnitude of the reflected radiant (light) flux from objects of a certain set {N} and depends on the wavelength of the radiant flux source, as well as the class of observed objects N = {A, B, C, ..., W}. It takes specific values in the range of values 0 <ρ (λ) <1.

Along with the choice of the number of registration zones, an important place is given to finding the optimal width of the registration zones Δλ i and their location in a given spectral region. The main requirement is to find such registration zones Δλ i in the spectral region of wavelengths from λ 1 to λ n , where for a larger number of objects, the maximum possible value of the spectral contrast value would be provided based on the reflectance of the objects [3].

In general, the effectiveness of using TV systems of spectral selection primarily depends on the possibility of choosing such the most informative areas of registration of the radiant flux from a certain set {M} = {Δλ 1 , Δλ 2 , ..., Δλ i , ..., Δλ m } that Satisfied the maximum difference of the observed objects by spectral-energy features. In this case, the following situations are possible for the observed space:

- the total number of objects equal to N is known, as well as the distribution of the spectral characteristics ƒ (λ) of these objects along the wavelength;

- there is general information about the number of objects and the course of their spectral characteristics along the wavelength;

- no reliable data (a priori uncertainty).

Each given situation requires a specific approach to the selection of radiant flux detection zones. So, for example, according to the reflectivity of the radiant flux, natural and artificial vegetation, cloud and snow cover, practically do not differ from each other in the VI region of the spectrum. At the same time, there are separate spectral zones of registration of the radiant flux Δλ i and Δλ j , where they differ from each other. Data on the spectral-energy characteristics of objects and general principles for the construction of optoelectronic systems have been duly reflected in the literature [4-6].

Known technical solutions and methods associated with the construction of spectrozonal TV systems [7-8]. The effectiveness of the proposed methods and technical solutions is directly related to the knowledge of the spectral characteristics of objects. Here, the choice of the registration areas of the radiant flux in the spectral region with a wavelength from Δλ i to Δλ n is required. providing the maximum difference of objects by spectral and energetic signs.

On the other hand, due to the possible low reflectivity of the objects and the magnitude of the reflected radiant flux towards the spectrozonal TV camera, when using “narrow” recording zones, there can be a low signal to noise ratio in the signals, which directly affects the accuracy of selection and recognition of objects. In addition, an increase in the number of generated signals of different spectral signals requires the organization of an identical number of input optical channels, and the use of multi-signal matrix photodetectors leads to a decrease in the resolution of the original images.

Of practical interest is the patent [9] for a method for the formation of spectrozonal signals, which are based on the integral-differential method of registering a radiant flux.

Using the integral method of recording by the input radiant flux and converting it into image signals allows you to switch to the differential method at the level of processing the zonal signals of TV images themselves. This method displays the integral-differential method of registering a radiant (light) flux and generating signals of different spectral images. This method uses zones of registration of a radiant flux with a variable width in the spectral range of wavelengths λ 1 to λ n .

The registration of the input radiant (light) flux with a variable width of the zone compared with narrow zones of registration of the radiant flux allows, on the one hand, to reduce the sensitivity requirements for TV sensors, or for a given sensitivity, allows to increase the signal-to-noise ratio in multispectral images.

The disadvantage of this method is that the total number of generated signals of different spectral images is still determined by the number of source signals with a variable width of the recording zone in the form

Figure 00000001

where m is the number of source signals of different spectral images, with a variable width of the registration area of the radiant flux.

As the closest analogue of the claimed invention according to the totality of the features and operations on the signals, the method of generating and displaying the spectrozonal television signals according to the patent [10] is adopted, which uses the integrated method of registering the radiant flux, where the radiant flux is recorded in a wide area of the optical spectrum, in comparison with the differential method, when the registration of the radiant flux is carried out in relatively narrow areas.

The essence of the formation and display of spectrozonal TV signals by a known method [10] is as follows. The formation of spectrozonal TV signals occurs according to a two-channel optical scheme, where the registration of the reflected radiant (light) or emitted stream is carried out within the entire wide spectral region with a wavelength of λ 1 to λ n , for which, after splitting the input radiant stream into two identical streams F (λ), it is passed through the PF 1 and PF 2 broadband optical filters having a spectral characteristic F 1 (λ) for the first TV sensor and the sensor for the second TV - F 2 (λ), wherein the spectral characteristics of ervogo and second optical filter covering the entire spectral region from λ 1 to λ n and satisfy F 1 (λ) = 1-P 2 (λ), and then carry out the division of the first signal to the second and form the resultant image signal in the form of U R ( t), then this signal is compared with reference signals UE , for example, comparing them in amplitude, then when these signals coincide, a selection signal U S (t) is generated for an object with a given known or unknown spectral characteristic in the observed space.

The use of optical filters, the spectral characteristic of which satisfies the condition Ф 1 (λ) = 1-Ф 2 (λ), allows you to generate integrated image signals and the resulting signal, which uniquely corresponds to an object having a specific spectral characteristic in the spectral region of wavelengths from λ 1 to λ n, thus allowing the automatic selection and implement the formation of finding information, the appearance or disappearance of the objects in the monitored area with a priori known or arbitrary n known spectral characteristics.

With this approach, the values of the resulting signal U R (t) can take specific values depending on the distribution of the spectral characteristics of objects in the range of values, namely

0 <U R (t) <1, or U R (t) = 1, and also 1 <U R (t) <T.

The reference amplitude values of the signals U e (t) with a given discreteness are in the range of values 0 <U e (t) <1, U e = 1 or 1 <U e (t) <D, while the value of the reference signal for the selection of the image signal of a particular the object is set in the range of values U E (min) <U E <U E (max) , which can be adjusted (decrease or increase).

The result of comparing the values of the signals U R (t) and U Э (t) is the formation of a selection signal U s (t), which carries information about the appearance, presence or disappearance of objects in the observed space, as well as subject to the selection of given objects from an arbitrary a priori known unknown spectral characteristics

The considered method does not allow the simultaneous formation of the amplitude values of the signals of spectrozonal images for wide and narrow zones of registration of the radiant flux in a given spectral region from λ 1 to λ n , that is, to combine the advantages of the integral and differential method of recording the radiant flux, which is its disadvantage.

EFFECT: provision of the possibility of generating signals of multispectral images for several zones of registration of the radiant flux within a wide spectral region and increasing the reliability of detection, selection and recognition of objects.

The technical result is achieved in that, in contrast to the known method for generating and displaying spectrozonal television signals, including recording the radiant (light) flux F (λ), inside a wide spectral region with a wavelength of λ 1 to λ n , splitting it into two identical streams F (λ) and their transmission through two broadband optical filters OF 1 and OF 2 , the spectral characteristic of which is mutually opposite and satisfies the condition Ф 1 (λ) = 1-Ф 2 (λ) in the spectral region with a wavelength from λ 1 to λ n, Conversion vanie radiant flux using two photodetector arrays having the same rectangular spectral response in the spectral portion having wavelength of λ 1 to λ n, and forming the two integral image signals U 1 (t) and U 2 (t), for which the first two integral the image signal U 1 (t) and U 2 (t) are modulated and the resulting resultant signal U S (t) is transmitted through the communication channel, the resulting signal U S (t) is demodulated at the receiving side and two original image integral signals U 1 (t ) and U 2 (t), more segregated of the first and second integral image signals U 1 (t) and U 2 (t) on the basis of their current amplitude values, to generate points in time Δt i, corresponding to one picture element i-m-signals of different spectral images

U 1 (Δλ1) (t), U 1 (Δλ2) (t), ..., U 1 (Δλi) (t), ... U 1 (Δλm) (t),

U 2 (Δλ1) (t), U 2 (Δλ2) (t), ..., U 2 (Δλi) (t), ..., U 2 (Δλm) (t),

by selecting them from a pre-prepared and stored data array in the memory device, which include a certain set of amplitude signal values for each recording zone Δλ 1 , with values distributed in the region 0≤ {U iΔλi (t)} ≤1, and then summarize separate signals of different spectral images corresponding to the amplitude value of the first and second integral image signal U 1 (t) and U 2 (t) and the formation of signals of different spectral images in the form

Figure 00000002

reflecting registration zones Δλ 1 , Δλ 2 , ... Δλ i , ... Δλ m , then the received m-signals of different spectral images U Δλ1 (t), U Δλ2 (t), ..., U Δλi (t), ..., U Δλm are processed (t), choose from the m-signals of different-spectrum images any three signals of different-spectrum images and feed them to the RGB inputs of a color video monitoring device and display them for visual analysis, as well as perform automatic selection of specified objects based on the analysis of the distribution of the amplitude values of m-signals of different-spectrum depicted and inside the spectral region with a wavelength of from λ 1 to λ n .

Using the integral registration method for the input radiant flux and converting it into image signals using optical filters, the spectral characteristic of which satisfies the condition Ф 1 (λ) = 1-Ф 2 (λ), allows, on the basis of the generated current amplitude value of two integrated image signals U 1 (t) and U 2 (t), reflecting the distribution of the spectral characteristics of the observed object i-be selected from a database prepared beforehand multispectral image signals Δλ1 U (t), U Δλ2 (t), ..., U Δλi (t), , U Δλm (t) for m-registration areas radiant flux, wherein m≥2.

Based on the spectral region with a wavelength from λ 1 to λ n , where the signals of the spectrozonal images are formed, it is possible to select the registration zones of the radiant flux with the same or arbitrary width. With the same width of the registration zone, their number can be found according to

m = (λ n1 ) / Δλ i .

With different widths of the registration zone, their number can be found from the condition

n1 ) = (Δλ 1 + Δλ 2 +, ..., + Δλ i +, ..., + Δλ m ).

To achieve this result, a method for generating, transmitting and reconstructing signals of different spectral images is proposed, including registration of the radiant (light) flux F (λ), inside a wide spectral region with a wavelength of λ 1 to λ n , splitting it into two identical flux F ( λ) and their transmission through two broadband optical filters OF 1 and OF 2 , the spectral characteristic of which is mutually opposite and satisfies the condition Ф 1 (λ) = 1-Ф 2 (λ) in the spectral region with a wavelength from λ 1 to λ n transformations radiant flux using two photodetector arrays having the same rectangular spectral response in the spectral portion having wavelength of λ 1 to λ n, and forming the two integral image signals U 1 (t) and U 2 (t), in which, first, two integral the image signal U 1 (t) and U 2 (t) are modulated and the resulting resultant signal U S (t) is transmitted via the communication channel, the resulting signal Us (t) is demodulated at the receiving side and two original integral image signals U 1 (t) are generated and U 2 (t), hereinafter separately for integral first and second image signals U 1 (t) and U 2 (t) on the basis of their current amplitude values, to generate the time Δt i of time corresponding to one picture element i-m-signals of different spectral images

U 1 (Δλ1) (t), U 1 (Δλ2) (t), ..., U 1 (Δλi) (t), ... U 1 (Δλm) (t),

U 2 (Δλ1) (t), U 2 (Δλ2) (t), ..., U 2 (Δλi) (t), ..., U 2 (Δλm) (t),

by selecting them from a pre-prepared and stored data array in the memory device, which include a certain set of amplitude signal values for each recording zone Δλ i with values distributed in the region 0≤ {U iΔλi (t)} ≤1, after which individual signals of different spectral images corresponding to the amplitude value of the first and second integral image signal U 1 (t) and U 2 (t) and the formation of signals of different spectral images in the form

Figure 00000003

reflecting registration zones Δλ 1 , Δλ 2 , ... Δλ i , ... Δλ m , then the received m-signals of different spectral images are processed U 1 (Δλ1) (t), U 1 (Δλ2) (t), ..., U 1 (Δλi ) (t), ... U 1 (Δλm) (t), choose any three signals of different spectral images from m-signals of different-spectrum images and feed them to the RGB inputs of a color video monitoring device and display them for visual analysis, as well as perform automatic selection of objects based on the analysis of the distribution of the amplitude values of m-signals of different spectral and siderations within the spectral portion having wavelength of λ 1 to λ n.

A spectral TV system that implements the proposed method for generating, transmitting, and reconstructing signals of different spectral images is shown in FIG. 1. Positions:

1 - lens;

2 - a device for splitting a radiant (light) stream into two identical streams;

3 - optical filters (hereinafter referred to as OF);

4 - control unit for optical filters;

5 - converters "radiant (light) flux-signal" (matrix photodetectors);

6 - a sync generator;

7 - amplifiers-shapers;

8 - signal modulator;

9 - communication channel;

10 - signal demodulator;

11 - block analysis, sampling and recovery of signals of different spectral signals;

12 - memory block signals of different spectral images;

13 - switch and distributor of signals of different spectral images;

14 - color video monitoring device;

15 - block automatic selection and recognition of objects;

16 - control unit.

If there is no need to transmit signals over the communication channel, then in the structural diagram of the system (Fig. 1) blocks 8, 9 and 10 will be absent.

The sync generator 6 generates the necessary horizontal and frame pulses, which are used to read images in matrix photodetectors (MFP) 5 1 and 5 2 , to form the output of the amplifiers-shapers 7 1 and 7 2 , the integrated image signals U 1 (t) and U 2 (t), and also enter the inputs of block 8.

As MFP 5 1 and 5 2 can be used CCD, CMOS photodetectors or other converters of the radiant flux into an electrical image signal. Since the original image signals are obtained by recording the radiant flux throughout in a wide spectral region with a wavelength from λ 1 to λ n , such signals will be integral image signals.

Optical filters (OF) 3 1 and 3 2 for the first and second channel of formation of integrated image signals U 1 (t) and U 2 (t) have spectral characteristics satisfying the condition Ф 1 (λ) = 1-Ф 2 (λ). Since the total number of PFs satisfying this condition can be different, separate groups of PFs are selected. To do this, using the OF 4 control unit, the spectral characteristics of the OF 3 1 and 3 2 are changed depending on their design (mechanical - by replacing the OF, or electronically).

In the spectrozonal TV system (Fig. 1), a two-channel optical scheme is used. Here, the total input radiant flux F (λ) is divided into two identical fluxes, each of which passes through its own optical filter having a spectral characteristic in accordance with the condition Ф 1 (λ) = 1-Ф 2 (λ).

After passing the first and second OF, the radiant flux F 1 (λ) and F 2 (λ) is projected onto the working surface of the first and second MFP 5 1 and 5 2 . Each formed integral signal of the image from their output goes to its input of the amplifier-former 7 1 and 7 2 , where the operations of amplification, separate signal processing and mixing with horizontal and frame pulses take place. From the output of block 7 1 and 7 2, the integrated image signals are fed to modulator 8.

When transmitting digital television signals through communication channels, various types of modulation are possible from amplitude manipulation (AMN) to multi-position quadrature amplitude manipulation (AMN) and its varieties. The following known types of modulation are possible:

- amplitude manipulation (AMN), consisting in a discrete change in the level of the amplitude of the carrier;

- frequency shift keying (FSK), consisting in a discrete change in the frequency of the carrier at a constant amplitude;

- phase shift keying (PSK), consisting in a discrete change in the phase of the carrier and its variants;

- multi-position quadrature amplitude manipulation (KAMN), etc.

The basis of modern methods for transmitting digital TV signals is the use of modulators and demodulators based on a quadrature scheme. Quadrature modulation is carried out using two balanced modulators, at the first inputs of which a transmitted signal U 1 (t) and U Q (t) are supplied, and a carrier with an offset phase of 90 ° is supplied to the second inputs. As is known, in case of balanced modulation in the case of the identity of both arms of the modulator, the signal at the output of the latter arises only when modulating signals are exposed. The output signal of the quadrature modulator is formed:

Figure 00000004

This signal U S (t) is transmitted through a communication channel 9, including a radio transmitting and receiving device, after which it is fed to a signal demodulator 10, at the output of which the original integral image signals U 1 (t) and U 2 (t) are generated.

It should be noted that synchronous detection is used to demodulate the signals. The correct operation of synchronous detectors depends on the frequency and phase of the carrier generated by the local generator and having synchronization with the transmitting part of the system. Quadrature modulation itself already provides a twofold increase in the efficiency of using the frequency band, since two signals are simultaneously transmitted on the same carrier frequency.

Further, the integrated image signals U 1 (t) and U 2 (t) are supplied to the inputs of the analysis, sampling and reconstruction unit 11. For the first and second integrated image signals U 1 (t) and U 2 (t), based on the analysis their current amplitude values are generated for a time point Δt i corresponding to one i-picture element of m-signals of different spectral images

U 1 (Δλ1) (t), U 1 (Δλ2) (t), ..., U 1 (Δλi) (t), ... U 1 (Δλm) (t),

U 2 (Δλ1) (t), U 2 (Δλ2) (t), ..., U 2 (Δλi) (t), ..., U 2 (Δλm) (t),

by selecting them from a previously prepared and stored data array in the memory device 12. Then, in block 11, the summation of the individual signals of different spectral images corresponding to the amplitude value of the first and second integral image signal U 1 (t) and U 2 (t) and the formation (restoration ) signals of different spectral images in the form

Figure 00000005

Thus, based on the analysis of the current amplitude values of the signals U 1 (t) and U 2 (t) at the output of block 11, m-signals of different spectral images are generated for each time point Δt i corresponding to one i-element of the image

U 1 (Δλ1) (t), U 1 (Δλ2) (t), ..., U 1 (Δλi) (t), ... U 1 (Δλm) (t),

with values distributed in the region 0≤ {U iΔλi (t)} ≤1. Then the m-signals of different-spectrum images are fed to the inputs of the switch-distributor of signals 13. In this block, any three signals of different-spectrum images can be selected in the case when their total number is m≥4 and is directed to the RGB inputs of a color video monitoring device (monitor). In addition, from the other outputs of the block 13 m-signals of different spectral images are fed to the inputs of the block automatic selection and recognition of objects 15.

The distribution characteristics of the spectral and energy characteristics of objects in the UV, VI, and IR spectral regions allow us to introduce the term “spectral portrait” of objects for their known set. Let us consider the features of the approach to determining the spectral portrait of objects (STR) and its possible application in spectral selection and recognition of objects in images [1].

The use of a certain set of zones {M}, including the registration zones of the radiant (light) flux Δλ 1 , ..., Δλ i , ..., Δλ m in the selected spectral region with a wavelength from λ 1 to λ n , and the formation of signals of different spectral images allows us to speak about open source software. It will be the distribution of the BOT of objects in a certain spectral region. Based on the analysis of the aggregate distribution of the values of the generated m-signals of different spectral images (analysis of STR), by performing arithmetic and logical operations on the signals, in block 15, automatic classification of objects {N} is carried out with their division into classes of objects N = {A, B, C , ..., S}.

It should be noted that the total number of generated amplitude values of m-signals of multi-spectral images satisfies the condition 2≤m≤P and is selected from a pre-prepared common array of signal data that includes individual data depending on the required number of generated signals of multi-spectral images. Their number can be equal to m = 2 or m = 3, or m = 4, ..., or m = P, which actually determine the width and the number of zones of registration of the radiant flux Δλ i , inside the spectral section with a wavelength from λ 1 to λ n .

From the control unit 16, control signals are sent to block 4 for changing the OF, to blocks 11, 12, 13, and 15 to control the analysis, sampling, switching, and recognition of signals from different spectral images. It should be noted that for the first and second channels for the formation of integrated image signals U 1 (t) and U 2 (t), the spectral characteristics of the used optical filters that satisfy the condition Ф i (λ) = 1-Ф j (λ) due to their shifts mechanically or electronically. I.e,

Consider a spectrozonal TV system (Fig. 1), which implements the proposed method for generating, transmitting, and reconstructing signals of different spectral images, using the example of observing objects with an arbitrary spectral characteristic.

Take for example the VI region of the spectrum from 380 nm to 760 nm, where the light flux is recorded. All visible colors are located in this part of the spectrum: violet (Δλ 1 = 380-430 nm), blue (Δλ 2 = 430-70 nm), blue (Δλ 3 = 470-500 nm), green (Δλ 4 = 500-560 nm), yellow (Δλ 5 = 560-590 nm), orange (Δλ 6 = 590-605 nm) and red (Δλ 7 = 605-760 nm). We write in general terms the expression for the generated integrated image signals in the spectral region with a wavelength from λ 1 to λ n

Figure 00000006

Figure 00000007

where k is the proportionality coefficient, τ (λ) is the transparency of the lens, ε 1 (λ) and ε 2 (λ) is the integral sensitivity of the first and second MFPs, ρ i (λ) is the reflection coefficient of the radiant (light) flux by the i-object, τ 1 (λ) and τ 2 (λ) are the transparency of the first and second lens in the spectral region from λ 1 to λ n .

We assume that ε 1 (λ) and ε 2 (λ) = 1, as well as τ (λ) = 1 in the spectral region of wavelengths from λ 1 to λ n . Let us also take, for example, that in these areas of the VI region of the spectrum, the reflectivity of some arbitrary class of objects A, B, C and the average transparency of the optical filters in the VI region of the spectrum for the individual recording zones were distributed as follows, as shown in Table 1.

Figure 00000008

Figure 00000009

In table 2, the values of the integrated image signals U 1 (t) and U 2 (t) were found according to

Figure 00000010
Figure 00000011

On the receiving side of the system, in the block for analysis and reconstruction of signals of different spectral images 11, based on the amplitude values of these signals, signals will be extracted from the memory unit 12

U 1 (Δλ1) (t), U 1 (Δλ2) (t), ..., U 1 (Δλi) (t), ..., U 1 (Δλm) (t),

U 2 (Δλ1) (t), U 2 (Δλ2) (t), ..., U 2 (Δλi) (t), ..., U 2 (Δλm) (t),

by selecting them from a previously prepared and stored data array in a memory device.

Further joint summation of the two signals (due to the use of OF 1 and OF 2 ) for each registration zone (table 2), for example, for a class A object in the registration zone Δλ 1 the value 0.8, for a class B object it will be 0, 8, and for an object of class C - 0.1. For the recording zone, for example, Δλ 5, there will be other signal values for an object of class A-0.8, for an object of class B-0.1 and for object C-0.6, etc., which will display the distribution of the spectral characteristics of objects (table 1).

Using the integral registration method for the input radiant flux and converting it into image signals using optical filters, the spectral characteristic of which satisfies the condition Ф 1 (λ) = 1-Ф 2 (λ), allows, on the basis of the generated current amplitude value of two integrated image signals U 1 (t) and U 2 (t) reconstruct signals of different spectral images U 1 (Δλ1) (t), U 1 (Δλ2) (t), ..., U 1 (Δλi) (t), ... U 1 (Δλm) ( t) for m-zones of registration of the radiant flux, with m≥2.

Literature

1. Sagdullaev Yu.S., Kovin S.D. Perception and analysis of multispectral images. M .: "Sputnik +", 2016. - 251 p.

2. Sagdullaev T.Yu., Sagdullaev Yu.S. On the choice of registration zones in spectrozonal television. Questions of radio electronics, ser. Television Engineering, 2011, no. 2, p. 3-25

3. Sagdullaev T.Yu., Sagdullaev Yu.S. Spectral selection and object recognition. Questions of radio electronics, ser. Technique of Television, 2012, issue. 2, pp. 96-106.

4. Krinov E.L. Spectral reflectivity of natural formations. - M.: Academy of Sciences of the USSR, 1947 .-- 168 p.

5. Grigoriev A.N., Oktyabrsky V.V. Indicators of informativeness of hyperspectral agents in solving problems of monitoring natural and anthropogenic processes. Military Space Academy named after A.F. Mozhaysky, St. Petersburg, 2013

6. Tarasov VV, Yakushenkov Yu.G. Multispectral Optoelectronic Systems / Special Technique, No 4, 2002

7. RF patent No. 2543985. The method of generating signals of television images of various parts of the spectrum / Kovin S.D., Sagdullaev Yu.S. - publ. 03/10/2015 Bul. Number 7

8. RF patent No. 2546982. The method of forming and displaying signals of color, spectrozonal and thermal imaging images / Kovin S.D., Sagdullaev Yu.S. g. - publ. 04/10/2015 Bul. Number 10

9. RF patent No. 2604898. The method of forming spectrozonal video signals / Kovin S.D., Sagdullaev Yu.S. - Priority 06/26/2015. Publ. 12/20/2016, Bull. Number 35

10. RF patent for the invention No. 2374783. The method of formation and display of spectrozonal television signals / Vilkova NN, Zubarev Yu.B., Sagdullaev Yu.S. - publ. November 27, 2009 Byul. No. 33.

Claims (6)

  1. A method for generating, transmitting, and reconstructing signals of multispectral images, including recording the radiant (light) flux F (λ) inside a wide spectral region with a wavelength of λ 1 to λ n , splitting it into two identical fluxes F (λ), and passing them through two broadband optical filters OF 1 and OF 2 , the spectral characteristic of which is mutually opposite and satisfies the condition Ф 1 (λ) = 1-Ф 2 (λ) in the spectral region with a wavelength from λ 1 to λ n , the conversion of radiant fluxes using two matrix fo of receivers having the same rectangular spectral characteristic in the spectral region with a wavelength of from λ 1 to λ n , and the formation of two integrated image signals U 1 (t) and U 2 (t), characterized in that at first two integrated image signals U 1 ( t) and U 2 (t) are modulated and the resulting resultant signal U S (t) is transmitted via the communication channel, on the receiving side, the resulting signal U S (t) is demodulated and two original integral image signals U 1 (t) and U 2 ( t), then separately for the first and second integral signal in image U 1 (t) and U 2 (t) on the basis of their current amplitude values to produce the time Δt i of time corresponding to one picture element i, m signals of different spectral images
  2. U 1 (Δλ1) (t), U 1 (Δλ2) (t), ..., U 1 (Δλi) (t), ..., U 1 (Δλm) (t),
  3. U 2 (Δλ1) (t), U 2 (Δλ2) (t), ..., U 2 (Δλi) (t), ..., U 2 (Δλm) (t),
  4. by selecting them from a pre-prepared and stored data array in the memory device, which include some set of amplitude values of the signals for each recording zone Δλ i , with values distributed in the region 0≤ {U iΔλi (t)} ≤1, and then summarize separate signals of different spectral images corresponding to the amplitude value of the first and second integral image signals U 1 (t) and U 2 (t), and the formation of signals of different spectral images in the form
  5. Figure 00000012
  6. reflecting registration zones Δλ 1 , Δλ 2 , ..., Δλ i , ..., Δλ m , then the received m signals of different spectral images U 1 (Δλ1) (t), U 1 (Δλ2) (t), ..., U 1 ( Δλi) (t), ..., U 1 (Δλm) (t), choose from m signals of different-spectrum images any three signals of different-spectrum images, feed them to the RGB inputs of a color video monitoring device and display them for visual analysis, as well as perform automatic selection specified objects based on the analysis of the distribution of the amplitude values of m signals of different spectral and siderations within the spectral portion having wavelength of λ 1 to λ n.
RU2018137166A 2018-10-22 2018-10-22 Method of forming, transmitting and restoring signals of different-spectral images RU2708454C1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
RU2018137166A RU2708454C1 (en) 2018-10-22 2018-10-22 Method of forming, transmitting and restoring signals of different-spectral images

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
RU2018137166A RU2708454C1 (en) 2018-10-22 2018-10-22 Method of forming, transmitting and restoring signals of different-spectral images

Publications (1)

Publication Number Publication Date
RU2708454C1 true RU2708454C1 (en) 2019-12-09

Family

ID=68836738

Family Applications (1)

Application Number Title Priority Date Filing Date
RU2018137166A RU2708454C1 (en) 2018-10-22 2018-10-22 Method of forming, transmitting and restoring signals of different-spectral images

Country Status (1)

Country Link
RU (1) RU2708454C1 (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6333757B1 (en) * 1993-11-12 2001-12-25 Reveo, Inc. Method and apparatus for producing and displaying spectrally-multiplexed images of three-dimensional imagery for use in stereoscopic viewing thereof
US6680739B1 (en) * 2000-11-17 2004-01-20 Hewlett-Packard Development Company, L.P. Systems and methods for compositing graphical data
US20120007979A1 (en) * 2008-04-16 2012-01-12 Elbit Systems Ltd. Advanced Technology Center Multispectral enhanced vision system and method for aircraft landing in inclement weather conditions
US20120062697A1 (en) * 2010-06-09 2012-03-15 Chemimage Corporation Hyperspectral imaging sensor for tracking moving targets
RU2543985C1 (en) * 2013-12-26 2015-03-10 Закрытое акционерное общество "МНИТИ" (сокращенно ЗАО "МНИТИ") Method of generating television image signals of different spectral regions
RU2604898C1 (en) * 2015-06-26 2016-12-20 Российская Федерация, от имени которой выступает Министерство обороны Российской Федерации Method of generating of multispectral video signals

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6333757B1 (en) * 1993-11-12 2001-12-25 Reveo, Inc. Method and apparatus for producing and displaying spectrally-multiplexed images of three-dimensional imagery for use in stereoscopic viewing thereof
US6680739B1 (en) * 2000-11-17 2004-01-20 Hewlett-Packard Development Company, L.P. Systems and methods for compositing graphical data
US20120007979A1 (en) * 2008-04-16 2012-01-12 Elbit Systems Ltd. Advanced Technology Center Multispectral enhanced vision system and method for aircraft landing in inclement weather conditions
US20120062697A1 (en) * 2010-06-09 2012-03-15 Chemimage Corporation Hyperspectral imaging sensor for tracking moving targets
RU2543985C1 (en) * 2013-12-26 2015-03-10 Закрытое акционерное общество "МНИТИ" (сокращенно ЗАО "МНИТИ") Method of generating television image signals of different spectral regions
RU2604898C1 (en) * 2015-06-26 2016-12-20 Российская Федерация, от имени которой выступает Министерство обороны Российской Федерации Method of generating of multispectral video signals

Similar Documents

Publication Publication Date Title
Lapray et al. Multispectral filter arrays: Recent advances and practical implementation
CN105282459B (en) Pixel sorting in imaging sensor
US20160170103A1 (en) Phase Gratings with Odd Symmetry for Optical Sensing
CN102447826B (en) Visible and infrared dual mode imaging system
US10182216B2 (en) Extended color processing on pelican array cameras
TWI489858B (en) Image capture using three-dimensional reconstruction
US8885073B2 (en) Dedicated power meter to measure background light level in compressive imaging system
Akkaynak et al. Use of commercial off-the-shelf digital cameras for scientific data acquisition and scene-specific color calibration
EP1952106B1 (en) System for multispectral imaging
CN103369339B (en) Formation method and equipment
US9420241B2 (en) Multi-spectral imaging
US7855786B2 (en) Single camera multi-spectral imager
US8553222B2 (en) Coded aperture snapshot spectral imager and method therefor
US7019777B2 (en) Multispectral imaging system with spatial resolution enhancement
US5258618A (en) Infrared imager forming two narrow waveband images of the same object field
JP5976676B2 (en) Imaging system using longitudinal chromatic aberration of lens unit and operation method thereof
CN100416843C (en) Imaging device and imaging method
CN101124462B (en) A system for multi- and hyperspectral imaging
EP1495293B1 (en) Imaging spectrometer
US7274393B2 (en) Four-color mosaic pattern for depth and image capture
US8411146B2 (en) Single camera color and infrared polarimetric imaging
US20170112376A1 (en) Systems and Methods for Lensed and Lensless Optical Sensing
US8619128B2 (en) Systems and methods for an imaging system using multiple image sensors
Waxman et al. Solid-state color night vision: fusion of low-light visible and thermal infrared imagery
US6781127B1 (en) Common aperture fused reflective/thermal emitted sensor and system