RU2553843C2 - Method for remote diagnostics of state of linear part of underground main pipelines - Google Patents

Abstract

FIELD: measurement equipment.

SUBSTANCE: method is intended for solution of a task of remote detection of precursors of emergency situations on underground main pipelines. The method is implemented by acquisition and analysis of images as per reflected and proper radiations of the underlying surface of the pipeline running route. Before the beginning of the route survey, there shaped are simulators of designated precursors with storage of their coordinates in a flight navigation device of a flying carrier. Images obtained during the flight are transformed to space of solutions by means of matched filters and by using as thresholds of taking decisions of output signals of filters from images of the corresponding simulators. At the same time, correlation functions of the received images are determined to calculate the number of false decisions and as per this number and the shaped space of solutions there determined is availability on the test route of precursors of emergency situations of the corresponding type.

EFFECT: improving reliability of detection, reducing the volume of information transmitted via a communication channel.

2 cl, 11 dwg

Description

The invention relates to the diagnosis of the state of existing underground trunk pipelines and can be used for reconnaissance of masked small targets on the underlying surface in the optical wavelength range.

A known method for the remote detection of oil leaks from the main pipeline [US Pat. 2073816 RF, IPC F17D 5/02. A method for remote detection of oil leaks from the main pipeline [Text] / Aleev P.M., Aleshko EN, Chepursky VN]. According to the method, an aerial survey of the thermal field of the pipeline route is carried out, local sections with an abnormal temperature are determined, laser probing of the underlying surface of the pipeline route at wavelengths of radiation absorption by the oil gas fraction components is carried out, and the leak location is determined by the location of the section with the anomalous temperature and increased absorption of the reflected laser radiation by gas fractions.

The disadvantage of this method is the limited number of detectable harbingers of emergency situations. In addition, the nature of the fluctuation of the thermal field of the pipeline route continuously changes over time and the appearance of a specific anomaly in the thermal image at the time of diagnosing the state of the investigated object is due to numerous reasons, one of which may be an oil leak. Therefore, the reliability of detection of this type of emergency cannot be high.

There is a method of detecting unauthorized taps in trunk pipelines [Shirobokov A.M. et al. Use of the Terma-2 multispectral thermal imager for monitoring oil trunk pipelines // Izv. universities. Instrumentation, 2002, t.45, No. 2, S.12-21]. It is proposed to use two ranges of the optical emission spectrum: 8–13.5 μm for detecting masked taps (“excavations”) of the stolen product and 1–1.2 μm for detecting oil spills and soils saturated with oil. Due to the commensurability of the signals from the detection objects with the standard deviation of the fluctuation of the background radiation intensity and a random change in the signal / noise ratio over time, it is not possible to provide an acceptable risk of decision-making in this method.

There is a method of searching for an object in the visible range of emissions, based on storing data on the reconnaissance area of the earth's surface and comparing these data with those recorded during the subsequent flight of an unmanned aerial vehicle [Application for US Pat. 2002114606 RF. Television method and search and video surveillance system [Text] / Prokofiev EV]. The disadvantage of this method is the instability of the results due to changes in the masking characteristics of the underlying surface in time, due to weather conditions and time of day.

A known method of monitoring the state of oil and gas pipelines, based on obtaining visible and thermal images of the prospected route, processing them to build a model of the flux density of thermal radiation and a model of block-fault structures, subsequent interpretation of the calculated data and mapping [Pat. 2428722 RF, IPC G01V 8/00. Method for remote diagnostics of trunk pipelines [Text] / Karimov K.M. and etc.]. The method is focused on the identification of large-scale changes in the pipeline laying zone and is not effective for detecting small-sized targets in its protection zone.

Of the known technical solutions, the closest in combination of essential features to the claimed is a method for diagnosing the condition of product pipelines [Application for US Pat. 2005110234/28 of the Russian Federation. IPC G01V 8/00. A method for diagnosing the condition of product pipelines [Text] / Baykov Yu.P. and etc.].

According to the method, the protection zone of the pipeline is surveyed in the visible and thermal ranges of radiation from the aircraft, the aircraft is positioned by flight and navigation tools, and the images obtained are interpreted. The disadvantage of this method is the neglect of the dynamic factor of the underlying surface parameters and the signal-to-noise ratio due to weather conditions and the time of shooting during the image interpretation operation. As a result, the reliability of detection of small targets in the known method is low.

The aim of the invention is to increase the reliability of decisions when interpreting thermal and visible images of the pipeline security zone for the presence of masked taps, undermines, leaks of the pumped product at any time and in any weather conditions.

This goal is achieved by the fact that simulators of emergency emergencies are formed on the route of the pipeline before carrying out the image acquisition, the coordinates of which are entered into flight and navigation tools, the obtained images are transformed into a decision space by means of filtering coordinated with the images and used as acceptance thresholds solutions of the output signals of the filters from the images of the corresponding simulators, at the same time determine the correlation functions and received images, followed by calculation by their parameters and the above decision thresholds of the potential number of false decisions, put marks on the decisions taken and the potential number of false decisions on the received images, transmit those decisions that were made on the radio station to the control center about the presence on them of manifestations of harbingers of emergency situations and which are provided for interpretation to a person or machine. In addition, if the next fragment of the image of the route exceeds the number of decisions made about the availability of emergency precursors of acceptable limits, only the coordinates of these fragments and the number of decisions made on them are transmitted to the control room, and if the upper limit is exceeded, the result of the interpretation of the transmitted data is reduced to the decision to re-diagnose of this section of the route, the lower border - about the need to eliminate detected harbingers of emergency situations.

при использовании согласованного фильтра для обнаружения квадратных объектов, фиг.9 - врезок соответственно для параметров ρ₁=ρ₂=0,75 (кривая 1), ρ₁=ρ₂=0,5 (кривая 2), ρ₁=ρ₂=0,25 (кривая 3), r_k - радиус корреляции флуктуации фона, S_a - площадь изображения согласованного фильтра.The invention is illustrated by the following description and the accompanying drawings. Figure 1 shows a graph of the temperature difference above the product outlet and away from it at a distance of 1 m, figure 2 shows the curve of the time variation of the standard deviation of the fluctuation of radiation of the underlying surface in the range of 8-14 μm on a sunny day, .3, 4 illustrates the change in the correlation function of thermal images of grass cover for one and a half hours, figure 5 presents the image of the site of the underlying surface, figure 6 is the result of its conversion by an aperture of the type “wall, parallel to the abscissa axis. " Fig.7 reflects the type of detected objects, Fig.8 - distribution of false emissions for the transformed field with a correlation function $$\chi (i,j)={\mathrm{<figure-callout\; id="2"\; label="\sigma "\; filenames="00000012.png,00000018.png"\; state="\{\{state\}\}">\sigma </figure-callout>}}^2{\rho }_{\mathrm{one}}^i\cdot {\rho }_2^j$$

when using a matched filter to detect square objects, Fig. 9 is an inset, respectively, for the parameters ρ ₁ = ρ ₂ = 0.75 (curve 1), ρ ₁ = ρ ₂ = 0.5 (curve 2), ρ ₁ = ρ ₂ = 0.25 (curve 3), r _k is the correlation radius of the background fluctuation, S _a is the image area of the matched filter.

при его преобразовании согласованным фильтром «круг», причем кривая 1 соответствует ρ=0,75, кривая 2 - ρ=0,5, кривая 3 - ρ=0,25.Figure 10 shows the change in the density of false decisions on the field with a correlation function $$\chi (x,y)=\sin \rho (\sqrt{x^2+{\mathrm{<figure-callout\; id="2"\; label="y"\; filenames="00000012.png,00000018.png"\; state="\{\{state\}\}">y</figure-callout>}}^2}/\rho \sqrt{x^2+y^2})$$

when it is converted by a matched “circle” filter, and curve 1 corresponds to ρ = 0.75, curve 2 - ρ = 0.5, curve 3 - ρ = 0.25.

, 10 - блок вычисления количества ложных решений на наблюдаемом изображении. На окончании фиг.11 - 11 - блок приема и запоминания переданной с летательного аппарата информации, 12 - монитор, 13 - оператор, 14 - кнопка (переключатель) принятия решений «Да», «Нет».Figure 11 presents a structural diagram that implements the proposed method: 1 - observed plot of the image of the track observed by an optical-electronic device, 2 - a database of images of detected emergency precursors, 3 - a block of matched filters, 4 - a block of threshold devices, 5 - a block for generating transmitted information, 6 - antennas, 7 - pilot-navigational aids, 8 - decision thresholds shaper, 9 - correlation function and parameter calculation unit $$S_a/\mathrm{<figure-callout\; id="2"\; label="\pi \; r\; l"\; filenames="00000012.png,00000018.png"\; state="\{\{state\}\}">\pi \; </figure-callout>}{r}_l^{}{}_2^{}$$

, 10 - block for calculating the number of false decisions in the observed image. At the end of Fig.11 - 11 is a block for receiving and storing information transmitted from the aircraft, 12 is a monitor, 13 is an operator, 14 is a decision button (switch) “Yes”, “No”.

The protection zone of the main pipeline is limited by “parallel vertical planes located on both sides of the axis of its linear part at a distance of 25 meters” [Technical regulation “On the safety of trunk pipelines for transporting liquid and gaseous hydrocarbons” / Project No. 408228-5, submitted by the Government of the Russian Federation] . Objects of detection: pits, taps, leaks of the pumped product. To install ammunition or tie-in, a pit is formed above the pipeline. The diameter of the pit is approximately 1 m. At the end of this work, a trench is excavated to divert the product abroad to the security zone (up to 500 m). The width of the trench is about 30-40 cm. A metal or plastic pipe with a diameter of 50-60 mm is laid in it. After this, the excavation sites are covered with soil and camouflaged by the removed turf. According to [Shirobokov A.M. et al. Use of the Terma-2 multispectral thermal imager for monitoring oil trunk pipelines // Izv. universities. Priborostroenie, 2002, vol. 45, No. 2, pp. 12-21] "the places of unauthorized taps are not visible to the eye either from the ground or from a patrol helicopter and can be detected by sequential piercing of soils with a metal rod. In the thermal range of optical radiation, a signal is recorded in the presence of a masked pit or insert.

Minor leaks of the pumped product (up to 50 l / h) are not recorded by traditional control methods based on measuring the pressure and flow rate of the product in the pipelines. There are emissions of the product into the atmosphere, soil, water, snow. The reflected emissions are recorded emissions of petroleum products into the atmosphere. Emissions to other media are recorded by thermal anomalies, which differ from the background radiation temperature by 0.6–6.8 K (in soil), 0.5–2 K (under snow cover). Images of thermal anomalies from leaks look mainly in the form of circles with “torn” edges [Aleev P.M., Ovsyannikov V.A., Chepursky V.N. Thermal imaging equipment for monitoring oil pipelines. - M .: Nedra, 1995. - 160 p.]. Given the existing sensitivity of thermal imaging systems (~ 0.05 K), solving the problem of air diagnostics of the state of trunk pipelines at first glance is not a difficult task.

The results of the analysis of the change in the amplitude of thermal anomalies of these species over time allowed us to conclude that it depends on weather conditions and physical parameters of the underlying surface [Egorov V.I. et al. Influence of local heterogeneity of soil on the temperature field of its surface / Scientific and Technical Bulletin of St. Petersburg State University ITMO, issue 31, 2006. - P.104-107.]. Figure 1 shows the experimental results of measurements of the temperature difference of the underlying surface above the place of the created product outlet and in the distance from it, obtained by us in the month of June. For a considerable time of the day, the place of removal is invisible in the thermal range of optical radiation.

Along with the signal amplitude, the camouflage characteristics of the underlying surface (the structure of the correlation function of radiation fluctuations) also change significantly in time. In the framework of the studies conducted by the authors, it was noted that at noon the correlation radius r _k (t) of these fluctuations is minimal, and the standard deviation σ _{q is} maximum. Over time, r _k (t) grows, σ _q decreases (thermal contrasts “blur”). The regression dependence of the indicated factors for grass cover according to the results of the experiments has the form

r _k (σ _q ) = r _k [σ _q (t ₀ )] - 0.11σ _q +0.1,

во времени дает представление график на фиг.2, о скорости изменения вида функции корреляции конкретного фона можно судить по графикам на фиг.3.where 0.11 is the normalizing factor of dimension, m ³ / W, t ₀ is the time of the onset of the isothermal regime, as defined in [Epifantsev B.N. On the estimated estimate of the dispersion of spatial fluctuations of the thermal radiation of the earth’s surface // Earth Exploration from Space, 1985, No. 3, pp. 40-45]. About the nature of the variance change $${\sigma }_q^{}$$$${}_2^{}$$

in time gives a view of the graph in figure 2, the rate of change in the form of the correlation function of a particular background can be judged by the graphs in figure 3.

Under conditions when the signal amplitude and masking characteristics of the background change in time according to an unknown law, a formalized statement of the problem of detecting emergency precursors, indicated above, is not possible. Tasks of this sort are entrusted to be solved by a person whose probability of detecting a target against the background of whose interference (empirical result) is described by the formula [Gushchin AB, Epifantsev B.N. An algorithm for detecting taps in underground pipelines on video images // Vestnik SibADI, issue 2 (20), 2011, p. 35-39]:

, $$R_{\mathrm{about}b.}=R_T\cdot R_{\Lambda }\cdot R_M=\left[\mathrm{one}-\exp \cdot \left(\frac{-C_b\cdot {\mathrm{TO}}^2<figure-callout\; id="3"\; label="\cdot \; \gamma "\; filenames="00000021.png,00000022.png"\; state="\{\{state\}\}">\cdot \; </figure-callout>{\gamma }^{}{}^3\cdot {\mathrm{AT}}_f^{0.3}\cdot {\tau }_c}{{(2\beta )}^2}\right)\right]$$

,

{1-exp [-0.15 (OSP-1) ² ]} · P _M

where P _T is the probability of detecting a single object on an ideal background (without interference), P _Λ is the probability of detecting objects of significant size depending on the signal-to-noise ratio (SIR), P _M is the probability component of detecting objects depending on the masking features of the underlying surface ( type of correlation function of the background), characterizing the forms of “spotting” of interference emissions, C _b = 16 deg ² · (cd / m ² ) ^-0.3 · (ang. min.) ^-3 · s ^-1 , K = ΔV / V _f - a luminance contrast of the object on the background brightness B _^, γ - the angular size of the object, τ _i = (L _E / υ _P) - time OISCA object, L _E - size observed on a monitor image area in the direction of flight of the screen, υ _P - exploration speed, 2β - the angular diameter of the field of view. The connection of the component P _M with the characteristics of the background fluctuation, as far as we know, has not been established. The principle of the formation of the decision-making threshold by the subject in solving problems of the indicated type is also not known.

However, the use of man as a decisive link in the system "Image acquisition system installed on the aircraft - the image transmission channel to the control center - operator (decoder)" is limited. First of all, the permissible speed of the decrypted image presented to the operator for the detection of small targets is not acceptable. Indeed, let us take the diameter of the pixel resolved above the underlying surface equal to 0.1 m (with a worse resolution, distinguishing a stationary person from a cobblestone becomes problematic). Then, when the expansion image by 600 lines represented on the monitor screen, we obtain _E L = 60 m. Taking SNR = 5, _p = 50 cd / m ^2, the screen size of 0.3 × 0.3 m, P _v = 0, 99, the distance "eye-screen" - 1 m, we obtain the following estimates. If K = 0.05, then υ _P = 30 km / h, with K = 0.1, υ _P <100 km / h /. These are unacceptable restrictions on flight speed, and the use of the "radiation converter - monitor - operator" system is productive when introducing a storage device for the received images and playing them at a speed controlled by the operator.

This technology has a significant drawback. So, the Land Sat satellite system gives every 18 days the most accurate photo of every square inch of the earth’s surface. However, due to the limitation of the rate of decryption by operators of 95% of the material received at the control point, “no one has ever seen, despite the urgent need to know what is happening to the Earth’s surface” [Hor. Email Earth on the scales. Ecology and the human spirit. - M .: PPP, 1993. - 432 p.]. The need to develop a new technology for aerial reconnaissance of long routes, based on the automation of the operation to detect abnormal deviations at the control object with the transfer of decryption results to the control point (control room), is obvious.

To date, there are no effective algorithms for detecting on the video paths of the underground pipelines of masked insets and places for the future output of the resulting products on the underlying surface.

The following solution to the problem is proposed.

Along the pipeline route, simulators of emergency precursors are formed:

masked pits and outlets of the pumped product, as well as a simulator of product leakage of a given intensity. The formation of the first two simulators does not have features, they implement the well-known approach to creating real branches. As for the latter option for creating a simulator, the effect of local cooling of the underlying surface should be used due to the more rapid exit to the surface of the gaseous component of the product [Aleev P.M., Ovsyannikov VA, Chepursky V.N. Thermal imaging equipment for monitoring oil pipelines. - M .: Nedra, 1995. - 160 p.]. In accordance with the Joule-Thomson law, when it reaches the surface (including the gas components of the soil displaced from the pores), its temperature decreases due to the pressure drop. There are several ways to simulate this effect. The simplest one is based on the creation of excessive pressure in the local area of the soil by generating neutral gas by chemical or physical methods.

Over time, due to a violation of the heat and mass transfer process in the soil-plant-air system, in the places where pits and product outlets are formed, an optical contrast is formed in the reflected radiation [Nerpin SV, Chudnovsky AF Energy and mass transfer in the plant-soil-air system. - L .: Gidrometeoizdat, 1975. - 360 p.]. The thermal contrast also changes. For example, for the range of 8-14 microns, radiation contrast can be represented as [Epifantsev B.N. Pipeline transport: neutralization of new security threats. - Omsk, SibADI Publishing House, 2006. - 295 p.]

, $${\mathrm{<figure-callout\; id="8"\; label="\mu "\; filenames="00000012.png,00000013.png"\; state="\{\{state\}\}">\mu </figure-callout>}}_8\approx {\left(\frac{\Delta \epsilon }{{\epsilon }_f}\right)}_8+\frac{\Delta T}{T_f-230}\cdot \left(\mathrm{one}-{\left(\frac{\Delta \epsilon }{{\epsilon }_f}\right)}_8\right)$$

,

where Δε and ΔT are the differences between the degrees of blackness and the temperature of the underlying surface in the area of the object and in the distance from it, 8 is the used radiation range of 8-14 microns. For well-masked product discharge trenches, radiation contrast may be absent: (Δε / ε _f ) = 0. Only the temperature component µ _Т = ΔТ / (Т-230) unmasks the object. After a certain time, the contrast component (Δε / ε _f ) increases and optical contrast appears in the reflected radiation in almost all the transparency windows of the atmosphere.

Exploration of the route is carried out using an aircraft.

In accordance with [Federal Law Technical Regulation “On the Safety of Trunk Pipelines for the Transport of Liquid and Gaseous Hydrocarbons” / Project No. 408228-5, introduced by the Government of the Russian Federation], the routes of trunk pipelines in the area should be indicated by columns installed at intervals of 1000 m and tied to the route ( coordinates).

You can take advantage of this circumstance and create simulators by tying their installation to the coordinates of the columns. Then, during the flight of the simulators, the pilot-navigational aids generate a signal, according to which decision-making levels are established for the forerunners of emergency situations indicated by the amplitude of the responses from the generated images of the simulators. According to preliminary estimates, the number of seats on a controlled route with a length of 250 km, where the installation of simulators is required, does not exceed 2.

The installation of simulators allows you to obtain information about signal levels from interesting precursors of emergency situations at the current time. At a low rate of change, these levels will correspond to the actual levels during the reconnaissance cycle (~ 1 h). Information of this kind is the basis to avoid missing a target (errors of the first kind). However, the issue of false alarms remains unclear. If you don’t know the density of false decisions, knowing only the threshold levels of useful signals does not allow us to consider the reconnaissance system of the described purpose effective.

Indeed, a comparison of the experimental curve of the standard deviation of the fluctuation of the radiation of the underlying surface in the range of 8-14 μm (Fig. 2) with changes in the useful signal (Fig. 1) leads to the conclusion that it is necessary to find ways to combat interference in this fragment of the application of signal detection theory.

The theory aims to use spatial filters consistent with the shape and geometric characteristics of the object, designed to increase the signal-to-noise ratio and thereby reduce the likelihood of false alarm [V. Levshin Spatial filtering in optical direction finding systems. - M.: Soviet Radio, 1971. 200 p.]. However, practice puts forward such high demands on the frequency of false decisions (one false alarm is considered normal for 350 hours of operation of the security system) [S. Zvezhinsky The problem of selecting perimeter detection tools // M .: Special Technique. - 2002. - No. 4. - S.36-41] that the use of such a conversion is insufficient to ensure them.

In search of a more effective solution, a number of computational experiments were carried out. Fragments of the path images were converted into other images by integrating the image values of the underlying surface at each point (in the order of a standard television scan) at the boundaries of the apertures (matched with the desired spatial filter object). As an example, figure 5, 6 shows the image of the source and converted fields.

Figure 7 shows the objects that needed to be detected in the image. The given indicator of the shape of object B is described in [Zhivichin A.N. Decryption of photographic images. - M .: Nedra, 1980. - 253 p.]

, $$B=\sqrt{P\overline{R}/{\mathrm{<figure-callout\; id="0"\; label="S"\; filenames="00000012.png,00000013.png"\; state="\{\{state\}\}">S</figure-callout>}}_0}$$

,

- средний радиус, определяемый по радиусам вписанной в контур (R_B) и описанный вокруг его (R₀) окружностей, S₀ - площадь объекта.where P is the perimeter along the contour of the object, $$\overline{R}=0.5(R_B+R_0)$$

- the average radius, determined by the radii inscribed in the circuit (R _B ) and described around its (R ₀ ) circles, S ₀ - the area of the object.

We introduce the density of false decisions

$$K_{\Lambda B}^{\text{'}\text{'}}=2K_{\Lambda B}\cdot S_a/S_P,$$

от вида корреляционной функции флуктуации изменения подстилающей поверхности, ее параметров и формы обнаруживаемых объектов, показаны на фиг.8, 9, 10.where s_a and S_P respectively the area of the aperture and the analyzed field, K_ΛB - the number of emissions exceeding the level of the standard deviation of the field fluctuation. Fragments of a computational experiment reflecting the dependence $$K_{\Lambda B}^{\text{'}\text{'}}$$

 from the type of the correlation function of fluctuations in the change in the underlying surface, its parameters and the shape of the detected objects are shown in Figs. 8, 9, 10.

Noteworthy is the strong dependence of the number of false decisions on the form of the correlation function of a random field, its parameters and the shape of the detected object. Without taking into account this information, it is impossible to reduce the frequency of false decisions at the output using consistent filtering even with the introduction of an adaptive decision threshold.

Based on the above results, the following method is proposed for solving the problem of detecting emergency precursors on the route of underground trunk pipelines.

The image of the route 1 in FIG. 2 during the flight is converted by matched spatial filters (sequential convolution of the images of the desired object with a fragment of the observed field at each point in the order of a television scan). This function is performed by blocks 2 and 3 (Fig. 11). When optical-electronic image converter simulators appear in the field of view, the coordinates of which are previously entered into the flight-navigation system 7, a signal is generated by which block 8 remembers the current levels from the converted images of the simulators (block 3 output) and which are used as decision thresholds before the emergence of another group of images of imitators of emergency precursors.

The first threshold device 4 outputs all the emissions of the matched filter, exceeding the set levels for each simulator. Block 5 assigns them the current coordinates and transmits the observed fragment of the track image with an indication (frame, arrow) of the detected places of emergency precursors and their coordinates.

, где r_k - интервал корреляции флуктуации поля. По этой информации и текущему порогу на выходе блока 8 блок 10 определяет число (плотность) ложных решений (алгоритм определения ясен при рассмотрении графиков на фиг.8, 9, 10) на анализируемом фрагменте изображения. Данная информация также передается на диспетчерский пункт. При отсутствии решений блока 4 о наличии искомых объектов в поле зрения системы разведки в канал посылается информационное сообщение на изображения с координатами x∈x_i, x_i+Δ₁, y∈y_i; y_i+Δ₂ объектов нет. Для оценки коэффициента сокращения объема посылаемой - принимаемой информации требуются трудоемкие эксперименты. В проведенных нами вычислительных экспериментах наблюдалось сокращение анализируемой информации на два порядка.Simultaneously with the above operations, in block 9, the form of the correlation function of the observed radiation (reflection) field and the exponent are determined $$S_a/\mathrm{<figure-callout\; id="2"\; label="\pi \; r\; k"\; filenames="00000012.png,00000018.png"\; state="\{\{state\}\}">\pi \; </figure-callout>}{r}_k^{}{}_2^{}$$

where r _k is the correlation interval of field fluctuations. Based on this information and the current threshold at the output of block 8, block 10 determines the number (density) of false decisions (the determination algorithm is clear when considering the graphs in Figs. 8, 9, 10) on the analyzed image fragment. This information is also transmitted to the control room. In the absence of decisions of block 4 about the presence of the desired objects in the field of view of the reconnaissance system, an information message is sent to the channel at the images with coordinates x∈x _i , x _i + Δ ₁ , y∈y _i ; y _i + Δ ₂ there are no objects. To assess the reduction coefficient of the amount of information sent and received, laborious experiments are required. In our computational experiments, a reduction of the analyzed information by two orders of magnitude was observed.

The receiving part of the system (antenna unit - storage device - monitor) works according to the following algorithm. The incoming information is stored in block 11 (end of FIG. 11). The operator 13 can display fragments of the stored images on the screen of the monitor 12 and make a decision that is consistent with the decision of the machine (arrow, frame) or not. Using buttons 14, he transmits his decision to block 11. In the first case, the analyzed image is transferred to the archive, in the second, the suspicious places on the highway and their coordinates are reported to the ground security service.

In contrast to the well-known decisions, the operator is provided with interpretation many times less information about the state of the trace for making a decision, time appears to examine the suspicious fragment of the trace image in more detail, and the reliability of detection increases.

The decision is made taking into account the decisions of the machine, working with the maximum possible amount of information about the characteristics of the desired objects and backgrounds at a given point in time, taking into account the dynamics of their parameters in the past. This approach eliminates the possibility of missing the target and informs the operator about the number of false decisions that accompany the correct decision. If their number is large (background emissions are similar to emissions from the desired objects) and exceed the accepted boundary level, unproductive decisions should be avoided by postponing the reconnaissance period to another time.

With a decrease in the number of false decisions, the probability of correct decisions quickly increases and when it reaches the established level of confidence, the operator does not hesitate to notify the security service about the need to eliminate the detected precursor of an emergency.

Since the signals from simulators obtained by reflected and intrinsic emissions may not be present on one channel and may be presented on another, the use of 2 channels for obtaining information leads to an increase in the reliability of detection of desired objects. In addition, the two-channel system allows us to solve the problem of establishing the time for emergence of emergency precursors on an operated underground pipeline.

Claims (2)

1. A method for remote diagnostics of the state of the linear part of underground trunk pipelines, based on the acquisition of images from the reflected and intrinsic emissions of the underlying surface of the pipeline security zone from the aircraft, its current positioning by flight and navigation means, image interpretation, characterized in that along the pipeline imaging work is formed by imitators of emergency harbingers, the coordinates of which are entered in navigational aids, obtained images are transformed into a decision space by means of filtering simulators compatible with images and using output signals of filters from images of corresponding simulators as decision thresholds, at the same time the correlation functions of the obtained images are determined, followed by calculation by their parameters and decision thresholds indicated above potential number of false decisions, mark the received decisions with marks on the decisions taken and A potential number of false decisions are transmitted over the air to the control room by those of these images on which decisions were made on the presence of signs of emergency precursors on them and which are provided for interpretation. 2. The method according to claim 1, characterized in that when the next image fragment exceeds the number of decisions made about the availability of emergency precursors of acceptable limits, only the coordinates of this image and the number of decisions made on it are transmitted to the control room, and the result of the interpretation of the transmitted data is reduced to the decision to re-diagnose this section of the route.

Images