WO2012000384A1 - 图像结冰探测器及探测方法 - Google Patents

图像结冰探测器及探测方法 Download PDF

Info

Publication number
WO2012000384A1
WO2012000384A1 PCT/CN2011/075787 CN2011075787W WO2012000384A1 WO 2012000384 A1 WO2012000384 A1 WO 2012000384A1 CN 2011075787 W CN2011075787 W CN 2011075787W WO 2012000384 A1 WO2012000384 A1 WO 2012000384A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
icing
marking
module
parameters
Prior art date
Application number
PCT/CN2011/075787
Other languages
English (en)
French (fr)
Inventor
陈迎春
叶林
张淼
葛俊锋
冯丽娟
刘铁军
周峰
Original Assignee
中国商用飞机有限责任公司
中国商用飞机有限责任公司上海飞机设计研究院
华中科技大学
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
Priority claimed from CN 201010219377 external-priority patent/CN102313512B/zh
Priority claimed from CN 201010219357 external-priority patent/CN102313510B/zh
Application filed by 中国商用飞机有限责任公司, 中国商用飞机有限责任公司上海飞机设计研究院, 华中科技大学 filed Critical 中国商用飞机有限责任公司
Priority to US13/807,964 priority Critical patent/US20130113926A1/en
Priority to EP11800129.6A priority patent/EP2589928B1/en
Publication of WO2012000384A1 publication Critical patent/WO2012000384A1/zh

Links

Classifications

    • BPERFORMING OPERATIONS; TRANSPORTING
    • B64AIRCRAFT; AVIATION; COSMONAUTICS
    • B64DEQUIPMENT FOR FITTING IN OR TO AIRCRAFT; FLIGHT SUITS; PARACHUTES; ARRANGEMENT OR MOUNTING OF POWER PLANTS OR PROPULSION TRANSMISSIONS IN AIRCRAFT
    • B64D15/00De-icing or preventing icing on exterior surfaces of aircraft
    • B64D15/20Means for detecting icing or initiating de-icing
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/0002Inspection of images, e.g. flaw detection
    • G06T7/0004Industrial image inspection
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/40Analysis of texture
    • G06T7/41Analysis of texture based on statistical description of texture
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30108Industrial image inspection
    • G06T2207/30156Vehicle coating

Definitions

  • the invention relates to an image water detecting device and a detecting method for obtaining information on the surface icing condition of an object by analyzing an image of the surface of the object, including whether it is icing, icing type, icing thickness and/or area. Wait. Background technique
  • icing conditions on specific surfaces or parts of an object need to be detected and analyzed.
  • the icing phenomenon of the leading edge of the wing tail, engine intake, etc. is monitored to prevent the icing from adversely affecting the flight and preventing icing from causing serious flight safety accidents.
  • ice as used in this application shall include all kinds of ice, frost and mixtures thereof.
  • earlier icing detection devices and methods include radiation, conductivity, and differential pressure.
  • the radiation icing detection device and method will bring great harm to human health, the reliability of the conductivity type icing detection device and method is poor, and the device used for the differential pressure type has a large volume and a complicated structure. , the response is slower.
  • these icing detection devices and methods can only give qualitative detection results of water condensation, and can not give quantitative information about icing thickness and water deposition rate.
  • the detector used in the magnetostrictive vibrating cylinder icing detection method is complicated in structure, high in production process, difficult to calibrate, and cannot be flush and conformally mounted on curved parts (such as the front wing of an aircraft wing).
  • the detection device used in the piezoelectric diaphragm type icing detection method although the volume and weight are small, can achieve a flush conformal installation of the curved surface portion to some extent, but the sensitive material is produced. The production requirements are stricter, the process is more complicated, and the assembly is more difficult.
  • fiber-optic icing detection still has several drawbacks: First, it cannot detect the supercooled Large Droplet Icing (SLD); secondly, it cannot eliminate or completely eliminate the type of icing. The effect on quantitative analysis; in addition, it can only achieve point detection, and it is impossible to detect faces with larger sizes.
  • SLD Large Droplet Icing
  • the object of the present invention is to provide a new image icing detector and detection method capable of processing an image of an ice layer by using image processing technology, thereby identifying differences between image features included in different water-splitting images, Based on these differences, the type of icing is accurately identified and data such as the thickness of icing is calculated.
  • an image icing probe comprising: an image acquisition system and an image processing system, wherein the image acquisition system is capable of acquiring an image of an object surface, and the image processing system is capable of analyzing the image, thereby Obtaining an icing condition on the surface of the object.
  • the determination of the icing condition is no longer obtained by means of a simple sensor signal, but by comprehensive information analysis of the entire image from multiple aspects, thereby greatly improving the qualitative detection of the identification icing type. And quantitative detection of water layer thickness, icing speed and/or area The accuracy.
  • the image processing system mainly comprises an icing analysis unit, which comprises a marking module, a calculation module and a judging module.
  • the marking module is configured to mark the image with a number of parameters related to an icing condition;
  • the calculating module is configured to calculate the marked parameter to obtain a characteristic factor of the image; And determining, according to the characteristic factor, an icing condition of the surface of the object.
  • the marking module performs marking based primarily on the brightness of the icing image, which is the most intuitive icing characteristic that reflects the icing condition.
  • the icing image There are always two effects of reflection and scattering in the ice layer. The reflection and scattering effects of different types of ice thickness are different. Differentiating the different brightnesses can well identify the corresponding reflection and scattering effects, and then identify the corresponding water conditions.
  • the marking module comprises a grayscale analysis module and/or a chromatographic analysis module, i.e., the image may be parameterized based on grayscale and/or chromatography.
  • the image processing system includes a point-taking module for acquiring at least a portion of the pixel points from the image for parameter marking by the marking module.
  • the points can be evenly distributed over the entire image, thus reducing the amount of data that needs to be processed. Different regions of the image can be distinguished according to requirements, and concentrated in important regions to take points, so that the detection results are more targeted.
  • the computing module calculates the feature factor based on the size and/or distribution of the marked parameters.
  • the surface image may be divided into a plurality of regions, and the feature factors are separately calculated for the respective regions. This allows the results of multiple regions to be compared to each other to avoid or reduce detection errors and errors.
  • the calculation module obtains the characteristic factor by a statistical method.
  • the range of values of the parameter marking by the marking module is divided into a plurality of intervals, and the distribution of the marked parameters within the intervals is taken as the characteristic factor.
  • the calculation module calculates the variance and/or the sum of the marked parameters and uses it as the characteristic factor.
  • the image processing system further includes an icing condition database.
  • the icing condition database includes feature data corresponding to various icing conditions for comparison with the feature factor.
  • the icing condition comprises an icing type, an icing thickness and/or an icing area.
  • the image processing system further includes an icing warning unit that determines whether the surface of the object is frozen.
  • the icing condition database includes a clean image of the surface of the object for comparison with the surface image.
  • an aircraft icing detector comprising an image icing detector in accordance with the first aspect of the invention.
  • the front end of the image acquisition system is disposed adjacent to the surface of the object to be detected for performing close-range microscopic detection of the surface of the object.
  • the front end of the image acquisition system is disposed away from the surface of the object to be detected for remote macroscopic detection of the surface of the object.
  • a method of detecting an icing condition on an object surface comprising the steps of:
  • the determination of the icing condition is no longer obtained by means of a simple sensor signal, but by comprehensive information analysis of the entire image from multiple aspects, thereby greatly improving the qualitative detection of the identification icing type. And the accuracy of quantitative detection of ice thickness, icing speed and/or area.
  • analyzing the image comprises:
  • the basis for marking the image with the parameters includes features of the image.
  • the reflection and scattering effects of different types of ice thickness are different, and the image characteristics reflected on the ice layer are also different. Differentiating different image features can correctly identify their corresponding reflection and scattering effects, and then identify the corresponding icing conditions.
  • the features of the image comprise brightness.
  • the parameter marking is performed by analyzing the gray scale and/or chromatogram of the image.
  • at least a portion of the pixels to be marked are acquired from the image prior to marking the image with the parameters.
  • the points can be evenly distributed on the entire image, thus reducing the amount of data that needs to be processed. It is also possible to distinguish different areas of the image according to the requirements, and focus on the important areas to take points, so that the detection results are more targeted. Sex.
  • marking the image with the image comprises comparing the image to a clean image when the surface of the object is not frozen, and using the contrasted result as a basis for the marking.
  • the basis for calculating the marked parameters includes the size and/or distribution of the marked parameters.
  • the labeled parameters are calculated using statistical methods.
  • the range of the marked parameters is divided into a plurality of intervals, and the distribution of the marked parameters within the plurality of intervals is taken as the characteristic factor.
  • the variance and/or sum of the marked parameters can be calculated and used as the characteristic factor.
  • the image is divided into a plurality of regions, and each of the regions is individually judged to be icing. This allows the results of multiple regions to be compared to each other to avoid or reduce detection errors and errors.
  • a database of icing conditions is provided for comparison with the images.
  • the icing condition of the surface of the object is obtained by comparing the characteristic factor with the characteristic data.
  • the database includes data regarding water species, water thickness and/or ice area.
  • the image icing detector and detection method of the invention can be widely applied to icing detection in various fields such as transportation, power equipment, field operation equipment and refrigeration equipment, and is also particularly suitable for icing detection application requirements of various aircrafts. Icing detection of different functions and requirements. DRAWINGS
  • FIG. 1 is a schematic diagram of an image acquisition system in an image icing detector according to a first preferred embodiment of the present invention
  • FIG. 2 is an image processing system in an image water detector according to a first preferred embodiment of the present invention.
  • Figure 3 is a schematic illustration of an image icing detector in accordance with a second preferred embodiment of the present invention, showing the arrangement of microscopic detection;
  • Figure 4 is a schematic illustration of an image icing detector in accordance with a third preferred embodiment of the present invention, showing the arrangement of macroscopic detection
  • Figure 5 is a schematic illustration of an image water jet detector in accordance with a fourth preferred embodiment of the present invention showing the arrangement of the detector for detecting from the side of the ice layer.
  • An image icing detector mainly comprises an image acquisition system and an image processing system, the former for acquiring an image from an object surface, and then the latter calculating and analyzing the acquired surface image, thereby finally Obtain the icing condition on the surface of the object.
  • FIG. 1 there is shown an image acquisition system 1-A of the image icing detector of the first preferred embodiment described above.
  • the core component of the image acquisition system 1-A is a imaging fiber bundle 104, which is capable of receiving a surface image of an object at the front end and transmitting the surface image along the optical fiber therein to other components connected to the rear end thereof.
  • the structure and principle of the imaging fiber bundle 104 itself are well known to those skilled in the relevant art and are not within the scope of the present invention.
  • the image fiber bundle has been widely used in many fields (such as gastroscope), so it will not be described here.
  • the advantage of using the imaging fiber bundle 104 in the present embodiment is that since the imaging fiber bundle 104 can achieve high-quality image propagation, the image of the surface of the object can be completely transmitted to a position away from the surface of the object, and finally Received by an image fixture disposed at a location remote from the surface of the object. This advantage is very important for some special applications.
  • the size of the space near the surface of the object to be detected for example, the wing of an aircraft
  • only equipment that meets the dimensional conditions is allowed to be installed.
  • existing imaging devices are difficult to meet this size requirement and thus cannot be applied.
  • the image fiber bundle only the front end of the very small size image fiber bundle can be placed near the surface of the object (for example, the wing), and the rear end thereof can be connected to the imaging device located away from the wing position, for example, Inside the aircraft. In this way, even larger existing imaging (including camera and video) devices can be used. According to the above method, the same effect as the miniaturization design can be achieved without downsizing the imaging device.
  • the environment of the surface of the object to be detected is relatively harsh.
  • the imaging device can be placed away from the surface of the object, and only the image fiber bundle remains in the vicinity of the surface of the object. Since the image fiber bundle itself is structurally simple and is not susceptible to damage, it can be easily applied to various detection environments and can protect a relatively vulnerable image forming apparatus.
  • the size of the imaging fiber used, the number of fibers, and the arrangement pattern of the fiber bundle can be reasonably determined according to different applications and implementations. Detailed description will not be given in this embodiment.
  • a focus lens 102 is coupled to the front end of the image fiber bundle 104 for receiving an image from the surface of the object.
  • the type thereof may be appropriately selected depending on the application form, for example, as will be described later, in the close-range microscopic detection, the head of the image fiber bundle 104 is very close to the surface of the object, and the focus lens 102 needs to use the lens. In the long-range macroscopic detection, the focus lens 102 needs to use a telephoto lens or a fisheye wide-angle lens.
  • a protective mirror 101 may also be disposed at the front end of the focus lens 102 to protect the focus lens 102 from the external environment, for example, to avoid abrasion of sand dust carried in a high-speed air stream on the surface of the object.
  • a coupling lens 107 and an image sensor 108 serving as image fixing means in the present embodiment are connected in series.
  • the coupling lens 107 can transmit the image collected by the focusing lens 102 and transmitted via the imaging fiber bundle 104 to the image sensor 108, and finally converted by the image sensor 108 into image information that can be recognized by the digital system for subsequent connection.
  • the image processing system (not shown in Figure 1) performs the analysis.
  • the image sensor 108 can include, for example, a CCD type or CMOS type image sensor as well as an infrared and/or ultraviolet image sensor to detect infrared rays and/or ultraviolet rays in the image of the surface of the object.
  • an electromagnetic wave emitting device 1 15 may be disposed near the front end of the image fiber bundle 104 for transmitting electromagnetic waves of a certain power and a certain wavelength band to the surface of the object, including visible light (400-760 nm) and infrared rays (760 nm to 1000 ⁇ m). And/or ultraviolet (1 to 400 nm), etc., and combinations thereof to achieve active detection.
  • the advantage is not only to overcome the adverse effects of detection when the ambient light is insufficient, but also to select certain specific wavelengths of electromagnetic waves as the detection source according to the detection requirements, so that it is particularly suitable for detecting ice of a specific type and a specific thickness range.
  • you can also Implementing composite probing makes the image information targeted in some applications more abundant.
  • the detection does not necessarily require an active signal source, and ambient light such as natural light is sufficient for the detection in many applications.
  • the active signal source does not need to be deliberately added, but can use existing equipment near the surface of the object.
  • the signal light on the surface of the aircraft can be used as the active signal source.
  • image information of different spectra can be selectively obtained by a filter (not shown) disposed between the protective mirror 101 and the image sensor 108.
  • an anti-icing and/or water removing device may be provided in the vicinity of the protective mirror 101 and the focus lens 102, for example, a occlusion (not shown) on the windward side and/or a micro
  • the electric heater 112 avoids and/or eliminates ice formed on the protective mirror 101 or the like, thereby eliminating the influence on the detection result.
  • An additional temperature sensor 111 can also be provided to prevent the heating temperature from being too high to damage the surface of the object or to image the fiber. On the other hand, temperature is an important reference for analyzing the icing condition.
  • the image acquisition system 1-A also includes some other accessory components, such as a flexible protective connector 105 that protects the imaging fiber bundle 104, a protective cover 106, and a connection line 116 that connects the electromagnetic wave transmitting device 115, and a power source required for the operation of the device. Lines and control signal lines, etc., are not described in detail.
  • control section mainly includes an image processing system 2-A, a temperature measurement and control system 2-B, a light source control system 2-C, and a central microprocessor 2-D.
  • the image processing system 2-A includes two parts: an icing warning unit 201, an icing analysis unit 202, and an icing condition database 203.
  • the icing warning unit 201 is specifically designed for image information processing in the initial stage of water sluicing, which can adopt high-speed image processing electronic system technology, can quickly obtain icing condition information during the initial icing period, and give an alarm signal to start icing. .
  • the shape of the probe is specially designed to make it easier to form water than the surface of the object to be detected, and it is possible to achieve an early warning before the surface of the object begins to freeze.
  • the working process of the icing warning unit 201 is: after receiving the image information transmitted by the image fixing device, comparing it with the clean ice-free image stored in the water condition database 203 without water, to determine whether Ice makes a judgment.
  • the specific determination process can be referred to the following description of the icing analysis unit 202.
  • the icing analysis unit 202 works in time with the icing warning unit 201 to enable qualitative and quantitative analysis of the specific effluent conditions (water sluice type, water sluicing thickness, and/or water sluicing area) of the surface of the object, It generally includes a parameter marking module, a calculation module, and a judgment module (not shown in the figure).
  • the icing analysis unit 202 Upon receiving the image information transmitted by the image fixing device, the icing analysis unit 202 first performs parameter marking on the image by the parameter marking module.
  • the marking method used may include gray processing and chromatographic processing, which are respectively implemented by a gray scale analysis module and a chromatographic analysis module, wherein the chromatographic analysis may further include analysis in a single color or multiple colors (for example, three primary colors).
  • the mark can be performed for all the pixels of the image, or by selecting a plurality of pixels from the point module, and multiple regions can be selected in the image and the average value of each region can be obtained, which mainly depends on Probing accuracy and speed requirements.
  • the parameters obtained by the marker are transmitted to the calculation module.
  • the function of the calculation module is to calculate the characteristic factors corresponding to the current surface image by calculating the received marker parameters, so that the subsequent judgment module can associate it with the icing condition database.
  • the feature data in 203 is compared.
  • the calculation method used may include, for example, statistics. Specifically, the value range of the parameter marking of the marking module may be divided into several intervals according to a predetermined criterion, and then the number of times all the marking parameters fall into each interval is counted, and Then get the percentage it occupies. In this case, the distribution of the marker parameters in each interval is the characteristic factor corresponding to the surface image of the object. Of course, it will be readily apparent to those skilled in the art that the division of the intervals can be taken in an uneven manner depending on the test results.
  • the function of the decision module is as described above for comparing the calculated feature factors with the existing feature data in the icing condition database 203 to find the feature data closest to the current feature factor.
  • the icing condition (including icing type, icing thickness, and/or icing area, etc.) corresponding to the characterization data can be considered as the current icing condition on the surface of the object.
  • a new reference amount such as the ambient temperature obtained by the temperature sensor 111, can also be introduced during the determination.
  • the icing condition database 203 is obtained by a large number of simulation tests and processing and classification of actual detection results. It can include several pieces of data, each of which includes a specific icing Status information (icing type, icing thickness, and/or icing area, etc.) and feature data corresponding to the icing condition, for the determination module to compare the feature data with the calculated feature factor to obtain a corresponding Icing condition.
  • icing Status information icing type, icing thickness, and/or icing area, etc.
  • the characteristics of the image such as the uniformity of the luminance values of the respective pixels (including the luminance of the gray and the luminance of the three primary colors).
  • the ice is ice, since the interior of the ice layer is approximately transparent, the electromagnetic waves reflected on the ice layer and the air interface can be received by the image fiber bundle with a large intensity, so the brightness values of the image pixels are large and both hook.
  • the reflection effect is greatly reduced due to the inclusion of air bubbles in the ice layer, and the scattering effect is strong, so the image pixel brightness value is low and unevenly distributed. Mixed ice is between the above two cases.
  • the brightness of the ice layer can also be examined. Because, after the icing type is determined, the thicker the icing thickness is within the thickness of a certain ice thickness, the greater the brightness of the image pixels.
  • the fetching point module in the icing analyzing unit 202 first selects a plurality of pixels (e.g., N) from above according to a predetermined rule.
  • the so-called rule means that the point can be taken only in a specific area of the image, or it can be relatively dense in some areas and relatively sparse in other areas.
  • the parameter marking module performs three primary color analysis on each selected pixel through common software in the field of image processing, and respectively obtains three primary color values of each pixel, thereby completing parameter marking. Taking an eight-bit microprocessor system as an example, each primary color has a value range of 0-255.
  • the three primary color values of the tag completion are transmitted to the calculation module.
  • each point is classified into its corresponding interval according to the previously divided three primary color value interval.
  • the division of each of the three primary color values can be performed as needed, for example, red, green, and blue light are respectively divided into p, q, and r segments, respectively.
  • the number n of points, n 2 , n 3 ... ⁇ ⁇ and the percentages m 2 , m 3 ... m K of the total number of points N are counted.
  • the number of points ⁇ n h n 2 , n 3 ... ⁇ ⁇ ⁇ or the percentage ⁇ x h x 2 , x 3 ... ⁇ ⁇ ⁇ is used as the feature factor corresponding to the current surface image.
  • the judging module compares the calculated feature factor with the feature data stored in the icing condition database 203, thereby selecting a piece of data in the database that is closest to the current icing condition, and using the data in the piece of data.
  • icing conditions freezing type, icing thickness, and/or icing area, etc.
  • the working process of the parameter marking module and the judging module is not too large.
  • different calculation methods are used in the calculation module.
  • the variance of the pixel brightness value can be used as the characteristic factor of the judgment, so that the distribution of the brightness can be more clearly seen; and when the thickness of the ice layer is examined, the sum of the brightness values of all the pixels can be used. As the final feature factor.
  • the quantitative detection result of the thickness of the icing can be conveniently obtained.
  • the above-described detector and detection method can also achieve a direct identification of the thickness of the ice layer observed on the side. As shown in Fig. 5, by identifying the grayscale and color in the image, the icing area and the surface area of the object can be clearly distinguished, and by analyzing the icing image, multiple measurement points are selected to determine the ice layer. The average thickness can also easily determine the thickness of the entire ice layer.
  • the temperature measurement and control system 2- ⁇ can acquire the temperature signal of the temperature sensor 111 and compare the temperature signal with the set temperature value to serve as a control basis for the operation of the heater 112. And can also The temperature value is transmitted to the judging module as a reference amount for judging the current icing condition.
  • the light source control unit 2-C controls the operation of the electromagnetic wave transmitting device 1 15, that is, controls the electromagnetic wave type, the emission time, and the transmission power.
  • the electromagnetic wave emitting device 115 can operate continuously or intermittently or irregularly. In the regular intermittent working state, the emission frequency of the electromagnetic wave can be selected as 1-20 Hz, which can work in coordination with the image acquisition system and ensure a sufficiently fast detection speed.
  • the operation of the light source control unit 2-C and the image processing system 2-A can be coordinated by the central microprocessor 2-D such that the image processing system 2 is only operated when the electromagnetic wave transmitting device 115 operates. -A only works.
  • the central microprocessor 2-D can realize the control of the image processing system 2-A, the temperature measurement and control system 2-B, and the light source control unit 2-C, and exchange information with each other, thereby realizing the functions of each unit.
  • the water jet detector and detection method according to the present invention can be used for microscopic detection of the water-splitting condition in a small area of the surface of the object.
  • the front end 110 of the image acquisition system is buried in the surface of the aircraft, and toward the outside, the image fiber bundle is led out from the front end 110 in the protective cover 106 and extends to an image processing system away from the front end (not shown). ).
  • the focus lens is a macro lens, and the image that can be acquired is limited to the very limited area that the front end is aligned with.
  • the distance is very close, it is possible to detect the image inside the ice layer, and the acquired image information is very accurate.
  • the resulting increase in icing thickness per unit time, i.e., the rate of icing is correspondingly more accurate.
  • the detector is mounted on the vertical tail of the aircraft and diagonally opposite the surface of the tail.
  • the focus lens is a telephoto lens or a fisheye wide-angle lens, and by adjusting the parameters of the lens, the entire surface area where icing detection is required (for example, a rectangular area of abcd shown in the figure) can be realized. Get a clear picture.
  • the detector In this arrangement, there is no icing at the front end of the detector, otherwise the icing image on the surface of the object to be detected will not be obtained. At this time, the setting of the anti-icing and de-icing device is very important. And since the image fiber bundle is a high temperature resistant glass fiber or a quartz fiber, as long as the temperature of the heating device is not too high, the detector will not be damaged.
  • This type of detection for a large area of the surface of the object has a lower accuracy in detecting the local point than the former, but an overall detection of the entire area can be achieved. From this perspective, the overall analysis accuracy of the icing condition can be improved. Because the uneven distribution of the ice layer may result in point detection and does not represent the overall icing condition, the conclusions based on several specific points deviate from the actual situation.
  • this form of macroscopic detection can produce special technical effects in certain special applications, such as the detection of "backflow ice” formed by large droplets of supercooling. This is very important for icing detection in areas such as aircraft.
  • So-called supercooled large water droplets refer to supercooled water droplets with a median volume diameter exceeding 50 microns. Since the supercooled large water droplets have a large mass, a large amount of latent heat needs to be released before the water is formed. It remains liquid for a period of time after contact with, for example, the surface of the aircraft, and no icing occurs. Only when the latent heat of the liquid is completely released, the icing occurs on a surface a certain distance backwards in the direction of the airflow. Therefore, in the case of "post-flowing ice", there may be a special case where ice is not formed at the leading edge portion of the aircraft wing and the empennage, and icing occurs at the non-protected portion after the leading edge.
  • the conventional detection method if such icing is to be detected, a large number of icing detector units are required to be placed on a large portion. In this way, not only is there a large installation space, but also damage to the structure of the surface of the object, and the provision of a plurality of icing detector units can also significantly increase the cost.
  • the detection of "post-flowing water" can be easily achieved. All that has to be done is to modify the algorithm in the calculation module (for example, to partition the surface of the object to be detected along the direction of the airflow, and calculate each region separately) so that it can achieve no icing on the surface of the object along the direction of the airflow. After the identification of the next icing condition, the post-flow ice detection can be realized.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Probability & Statistics with Applications (AREA)
  • Quality & Reliability (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)

Description

图像结冰探测器及探测方法 技术领域
本发明涉及一种图像结水探测装置及探测方法, 用于通过对物体表面的 图像进行分析而获得物体表面结冰状况的信息, 包括是否结冰、 结冰类型、 结冰厚度和 /或面积等。 背景技术
在很多情况下, 需要对物体的特定表面或部位的结冰状况进行探测和分 析。 例如在寒冷地区, 需要对冬季公路路面的结冰状况进行监测, 需要对风 力发电机的叶片及部分转动部件的结冰状况进行探测, 以及在飞机飞行过程 中, 对机体的多部位(如风挡、 机翼尾翼前缘、 发动机进气道等) 的结冰现 象进行监测, 以避免结冰对飞行造成不利影响, 防止结冰导致严重的飞行安 全事故。 需要注意的是, 本申请中所涉及的词语 "冰" 应当包括各种冰、 霜 及其混合物。
迄今为止, 人们已经设计和制造了多种用于结冰探测的装置并提出了多 种用于结冰探测的方法, 从而可以采取相应的措施来避免结冰危害的发生。 但是, 这些已有的结冰探测装置及方法均存在各种缺陷和不足, 从而极大地 影响了其性能和适用范围。
例如, 较早的结冰探测装置及方法包括放射线式、 电导率式和差压式。 其中, 放射线式结冰探测装置及方法会给人体健康带来很大的危害, 电导率 式结冰探测装置及方法的可靠性较差, 差压式所用的装置的体积较大, 结构 比较复杂, 响应速度较慢。 此外, 这几种结冰探测装置及方法均只能给出结 水与否的定性探测结果, 而不能给出关于结冰厚度和结水速率的定量信息。
现在广泛应用的是磁致伸缩振动筒式和压电膜片式结冰探测器及方法, 它们均能够给出一定冰厚范围内结水厚度和结冰速率的定量信息。但它们也 各有一定的缺陷: 磁致伸缩振动筒式结冰探测方法所用的探测器结构复杂、 生产工艺要求高、 校准困难, 并且无法齐平保形地安装于曲面部位(如飞行 器机翼尾翼前缘); 压电膜片式结冰探测方法所用的探测装置虽然体积、 重 量较小, 能够一定程度上实现曲面部位的齐平保形安装, 但其敏感材料的生 产要求较严, 工艺较为复杂, 装配比较困难。
近年来, 研究人员又提出了新的光纤式结冰探测器及探测方法, 其具有 一些突出的优点, 例如, 探测灵敏度高、 结构简单、 可靠性高以及能够实现 齐平保形安装等, 同时也具备一定的对传统结冰类型 (明冰、 淞冰以及混合 型冰) 的识别能力。 但是, 光纤式结冰探测仍然具有几方面的缺陷: 首先, 其无法实现对过冷大水滴结冰 ( Supercooled Large Droplet Icing, 筒称 SLD ) 的探测; 其次, 其无法消除或完全消除结冰类型对定量分析的影响; 此外, 其只能实现点探测, 而无法对具有较大尺寸的面进行探测。
对于过冷大水滴结冰的探测, 现有技术中存在一些尝试。 例如, 在公开 号为 US2002/0158768 A1 和 US2004/0231410 A1 的美国专利和申请号为 PCTAJS012106 的国际专利申请中, 都公开了一种能够实现过冷大水滴结冰 探测的装置。 但是, 这些探测器的核心传感和探测元件仍然采用磁致伸缩谐 振式和压电式结冰传感器, 因而也不可避免地具有前面提到的现有结冰探测 器的共同缺陷。
对于结冰类型的准确识別, 虽然现有的探测器能够一定程度上实现对结 冰类型的判断, 但是其准确度均比较低。 而要对结冰进行精确的定量分析, 必须建立在对结冰类型准确识别的基础上。 因为在现有的结冰探测装置中, 不同厚度的冰可能因为其类型不同而对应于同一输出信号。 发明内容
本发明的目的在于提出一种新的图像结冰探测器及探测方法, 其能够利 用图像处理技术对冰层的图像进行处理, 从而识别出不同结水图像所包含的 图像特征之间的差别, 并且根据这些差别准确地识别出结冰种类并计算结冰 的厚度等数据。
根据本发明的一个方面, 提出一种图像结冰探测器, 包括: 图像获取系 统和图像处理系统, 其中, 图像获取系统能够获取物体表面的图像, 图像处 理系统能够对所述图像进行分析, 从而得到所述物体表面的结冰状况。
通过上述技术方案, 结冰状况的判定不再借助于简单的传感器信号, 而 是通过对整幅图像从多个方面进行综合性信息分析而得出, 因而可以大大提 高识别结冰种类的定性探测和探测水层厚度、结冰速度和 /或面积的定量探测 的精确度。
优选地, 所述图像处理系统主要包括结冰分析单元, 其包括标记模块、 计算模块和判断模块。 其中, 所述标记模块用于对所述图像标记与结冰状况 相关的若干参数; 所述计算模块用于对所标记的参数进行计算, 以获得所述 图像的特征因数; 所述判断模块用于根据所述特征因数判断得到所述物体表 面的结冰状况。
优选地, 所述标记模块进行标记的依据主要是结冰图像的亮度, 这是能 够反映结冰状况的最直观的结冰特性。 冰层中总是同时存在反射和散射两种 效应, 不同种类不同厚度的结冰其反射和散射效应都是不相同的。 对不同亮 度进行区分即可很好地识别出其相应的反射和散射效应, 并进而识别出相应 的结水状况。
优选地, 所述标记模块包括灰度分析模块和 /或色谱分析模块, 即可以灰 度和 /或色谱为依据对图像进行参数标记。
优选地, 所述图像处理系统包括取点模块, 用于从所述图像上获取至少 一部分像素点以通过标记模块进行参数标记。取点既可以是均匀分布在整个 图像上, 从而减少需要处理的数据量, 也可以根据需求对图像的不同区域进 行区别, 集中在重要区域取点, 从而使得探测的结果更加具有针对性。
优选地, 所述计算模块计算所述特征因数的依据是所标记参数的大小和 /或分布。
优选地, 可以将表面图像划分为多个区域, 并对各个区域分别计算特征 因数。 这样可以使得多个区域的结果相互之间进行比较, 以避免或减少探测 的错误和误差。
优选地, 所述计算模块通过统计的方法获得所述特征因数。
优选地, 标记模块进行参数标记的取值范围被划分为多个区间, 并将所 标记的参数在这些区间内的分布作为所述特征因数。
优选地, 所述计算模块计算所标记参数的方差和 /或总和, 并将其作为所 述特征因数。
优选地, 所述图像处理系统还包括结冰状况数据库。
优选地, 所述结冰状况数据库中包括与各种结冰状况相对应的特征数 据, 用于和所述特征因数进行对比。 优选地, 所述结冰状况包括结冰种类、 结冰厚度和 /或结冰面积。
优选地, 所述图像处理系统还包括对所述物体表面是否结冰进行判断的 结冰预警单元。
优选地, 所述结冰状况数据库包括所述物体表面的清洁图像, 用于和表 面图像进行对比。
根据本发明的另一个方面, 提出一种飞行器结冰探测器, 其包括根据本 发明第一个方面的图像结冰探测器。 '
优选地, 图像获取系统的前端靠近待探测的物体表面而设置, 用于对所 述物体表面进行近距离的微观探测。
优选地, 图像获取系统的前端远离待探测的物体表面而设置, 用于对所 述物体表面进行远距离的宏观探测。
根据本发明的再一个方面, 提出一种探测物体表面结冰状况的方法, 包 括以下步骤:
- 获取待探测的物体表面的图像,
-对所述图像进行分析, 从而得到所述物体表面的结冰状况。
通过上述技术方案, 结冰状况的判定不再借助于简单的传感器信号, 而 是通过对整幅图像从多个方面进行综合性信息分析而得出, 因而可以大大提 高识别结冰种类的定性探测和探测冰层厚度、结冰速度和 /或面积的定量探测 的精确度。
优选地, 对所述图像进行分析包括:
- 给所述图像标记与结冰状况相关的若干参数,
-对所标记的参数进行计算, 从而获得所述图像的特征因数,
-根据所述特征因数得到所述物体表面的结冰状况。
优选地, 给所述图像标记参数的依据包括图像的特征。 冰层中总是同时 存在反射和散射两种效应, 不同种类不同厚度的结冰其反射和散射效应都是 不相同的, 反映到冰层上的图像特征也各不相同。 对不同图像特征进行区分 可以 艮好地识别出其相应的反射和散射效应, 并进而识别出相应的结冰状 况。
优选地, 所述图像的特征包括亮度。
优选地, 所述参数标记通过对图像的灰度和 /或色谱进行分析而进行。 优选地, 在给所述图像标记参数之前, 先从所述图像上获取至少一部分 待标记的像素点。 取点既可以是均勾分布在整个图像上, 从而减少需要处理 的数据量, 也可以才艮据需求对图像的不同区域进行区别, 集中在重要区域取 点, 从而使得探测的结果更加具有针对性。
优选地, 给所述图像标记参数包括将所述图像与所述物体表面未结冰时 的清洁图像进行对比, 并将对比后的结果作为标记的依据。
优选地, 对所标记的参数进行计算的依据包括所标记的参数的大小和 / 或分布。
优选地, 采用统计的方法对所标记的参数进行计算。
优选地, 将所标记参数的范围划分为多个区间, 并将所标记的参数在所 述多个区间内的分布作为所述特征因数。
优选地,可以计算所标记参数的方差和 /或总和, 并将其作为所述特征因 数。
优选地, 将所述图像划分为多个区域, 并对各个区域分别判断其结冰情 况。 这样可以使得多个区域的结果相互之间进行比较, 以避免或减少探测的 错误和误差。
优选地, 提供结冰状况数据库, 用于和所述图像进行对比。 通过将所述特征因数与所述特征数据进行对比, 从而得到所述物体表面的结 冰状况。
优选地,所述数据库包括关于结水种类、结水厚度和 /或结冰面积的数据。 本发明的图像结冰探测器及探测方法可广泛应用于运输、 电力设备、 野 外作业设备和制冷设备等多种领域的结冰探测, 还特别适合于各种飞行器的 结冰探测应用需求, 实现不同功能和要求的结冰探测。 附图说明
在对本发明的实施方式进行详细描述的过程中, 将参照下列附图: 图 1是根据本发明第一优选实施方式的图像结冰探测器中的图像获取系 统的示意图;
图 2是根据本发明第一优选实施方式的图像结水探测器中的图像处理系 统的示意图;
图 3是根据本发明第二优选实施方式的图像结冰探测器的示意图, 其中 显示了微观探测的布置方式;
图 4是根据本发明第三优选实施方式的图像结冰探测器的示意图, 其中 显示了宏观探测的布置方式;
图 5是根据本发明第四优选实施方式的图像结水探测器的示意图, 其中 显示了探测器从冰层侧面进行探测的布置方式。 具体实施方式
下面将参照附图来对本发明的优选实施方式进行详细描述。
根据本发明第一优选实施方式的图像结冰探测器主要包括图像获取系 统和图像处理系统, 前者用于从物体表面获取图像, 然后由后者对所获取的 表面图像进行计算和分析, 从而最终得到物体表面的结冰状况。
首先参照图 1 , 其显示的是上述第一优选实施方式的图像结冰探测器的 图像获取系统 1-A。
其中, 图像获取系统 1-A的核心部件是传像光纤束 104, 其能够在前端 接收物体的表面图像, 并将该表面图像沿着其中的光纤传输至与其后端相连 的其它部件。 传像光纤束 104本身的结构和原理已被相关领域的技术人员所 熟知, 并不属于本发明的范围。 并且作为一个成熟的技术, 传像光纤束已经 在很多领域(例如胃窥镜) 中得到了广泛的应用, 因此这里不再赘述。
而在本实施方式中采用传像光纤束 104 的优势在于, 由于传像光纤束 104能够实现对图像的高质量传播, 因此可以将物体表面的图像完整地传输 至远离物体表面的位置, 并最终由布置在远离物体表面位置的图像固定装置 接收。 这一优势对于某些特殊的应用来说是非常重要的。
在一个例子中, 对待探测的物体表面附近的空间尺寸有严格要求(例如 飞行器的机翼) , 此时只有满足尺寸条件的设备才被允许安装。 而现有的成 像设备很难满足这一尺寸要求, 因而无法得到应用。 但是通过传像光纤束, 可以仅将尺寸非常小的传像光纤束的前端设置在物体表面(例如机翼)附近, 而将其后端连接至位于远离机翼位置的成像设备上, 例如位于机航内部。 这 样, 即使是现有的体积较大的成像(包括照相和摄像)设备也能得到应用。 通过上述方法, 在未对成像设备进行小型化设计的情况下, 即可实现与小型 化设计相同的效果。
而在另一个例子中, 待探测的物体表面的环境比较恶劣。 此时通过传像 光纤束, 可以使得成像设备远离物体表面而设置, 而在物体表面附近仅保留 传像光纤束。 由于传像光纤束本身结构筒单, 不容易发生损坏, 因此可以方 便地适用于各种探测环境, 并且可以保护对相对易损坏的成像设备。
在实际应用中, 所釆用的传像光纤的尺寸、 光纤数量以及光纤束的排列 模式, 可以根据不同应用场合和具体实施而合理地确定。 本实施例中不再进 行详细的描述。
聚焦镜头 102连接在传像光纤束 104的前端,用于从物体表面接收图像。 其种类可以根据应用形式的不同而合适地选择, 例如下文将描述的, 在近距 离的微观探测中, 传像光纤束 104的头部非常靠近物体表面, 此时聚焦镜头 102需要选用 ί距镜头; 而在远距离的宏观探测中, 聚焦镜头 102就需要选 用长焦镜头或是鱼眼广角镜头。
在聚焦镜头 102的前端还可以设置有保护镜 101, 以保护聚焦镜头 102 免受外界环境的损害, 例如, 避免位于物体表面的高速气流中携带的砂尘的 磨损。
在传像光纤束 104后端, 依次串联有在本实施方式中用作图像固定装置 的耦合镜头 107和图像传感器 108。 其中, 耦合镜头 107能够将由聚焦镜头 102汇聚并经由传像光纤束 104传输的图像传输给图像传感器 108, 并最终 由图像传感器 108转换为可供数字系统识别的图像信息, 以供其后连接的图 像处理系统(图 1中未显示)进行分析。 图像传感器 108可以才艮据实际情况 包括例如 CCD型或是 CMOS型图像传感器,也可以包括红外和 /或紫外图像 传感器, 从而对物体表面图像中的红外线和 /或紫外线进行探测。
此外, 还可以在传像光纤束 104的前端附近设置电磁波发射装置 1 15, 用于向物体表面发射一定功率和一定波段的电磁波, 包括可见光(400-760 纳米) 、 红外线 ( 760纳米至 1000微米)和 /或紫外线 (1至 400纳米)等, 以及它们的组合, 以实现主动探测。 其优势不仅在于可以克服环境光不足时 给探测带来的不利影响, 还能根据探测需要选用某些特定波段的电磁波作为 探测源, 从而特别适用于探测特定类型和特定厚度范围的冰。 此外, 也可以 实现复合探测, 使得在某些应用中所针对的图像信息更为丰富。 当然, 本领 域技术人员能够理解的是, 探测并不一定要借助主动信号源, 自然光等环境 光在很多应用中就足以满足探测的要求。 并且在有些应用中, 主动信号源也 不需要故意添加, 而是可以利用物体表面附近已有的设备, 例如在飞行器应 用中, 可以利用飞行器表面的信号灯作为主动信号源。
同时, 通过设置在保护镜 101和图像传感器 108之间的滤光片 (图中未 显示) , 可以有选择地获得不同光谱的图像信息。
进一步地, 为了确保探测的顺利进行, 还可以在保护镜 101和聚焦镜头 102的附近设置防冰和 /或除水装置, 例如设置在迎风面上的遮挡(图中未显 示 )和 /或微型电加热器 112, 以避免和 /或消除在保护镜 101等上形成的冰, 从而排除对探测结果的影响。 还可以设置额外的温度传感器 111 , 一方面可 以防止加热温度过高而损坏物体表面或传像光纤, 另一方面温度是对结冰状 况进行分析的重要参考。
图像获取系统 1-A还包括一些其它的附属部件, 例如保护传像光纤束 104等的柔性保护接头 105、保护套 106, 和连接电磁波发射装置 115的连接 线 116, 以及设备工作所需的电源线和控制信号线等, 均不再进行详述。
下面参照图 2, 其显示的是才艮据本发明的第一优选实施方式的图像结冰 探测器的控制部分。 如图 2所示, 控制部分主要包括图像处理系统 2-A、 温 度测控系统 2-B、 光源控制系统 2-C以及中央微处理器 2-D。
其中, 图像处理系统 2-A包括二部分: 结冰预警单元 201、 结冰分析单 元 202和结冰状况数据库 203。
结冰预警单元 201专门针对结水初始期的图像信息处理, 其可以采用高 速图像处理电子系统技术, 能在结冰初始期快速地获得结冰状况信息, 并给 出开始发生结冰的报警信号。 如果配合探头一起使用, 其中探头的形状经过 特殊设计使其比待探测的物体表面更容易结水, 还能达到在物体表面开始结 冰之前即提前进行预警的效果。
结冰预警单元 201的工作过程为, 在接收到由图像固定装置传输的图像 信息后, 将其与结水状况数据库 203中储存的未结水时的清洁无冰图像进行 对比, 以对是否结冰做出判断。 其具体的判断过程可参考下面对于结冰分析 单元 202的描述。 结冰分析单元 202在时间上与结冰预警单元 201并行地工作, 能够对物 体表面的具体结水状况(结水种类、 结水厚度和 /或结水面积)做出定性和定 量分析, 其大致包括参数标记模块、计算模块和判断模块(图中均未显示)。
在接收到由图像固定装置传输的图像信息后, 结冰分析单元 202首先通 过参数标记模块对图像进行参数标记。 所用的标记方式可以包括灰度处理和 色谱分析处理, 分别通过灰度分析模块和色谱分析模块实现, 其中, 色谱分 析处理中又可包括采用单色或多种颜色(例如三基色)进行分析。 并且, 标 记既可以针对图像的所有像素点进行, 也可以通过取点模块从中选取若干个 像素点进行, 还可以在图像中选取多个区域并求得每个区域的平均值, 这主 要取决于探测精度和速度的要求。 在完成对表面图像的参数标记后, 标记获 得的参数将被传输至计算模块。
计算模块的作用在于, 通过对接收到的标记参数进行计算, 由此得出当 前表面图像所对应的特征因数 , 以供后续的判断模块将其与结冰状况数据库
203中的特征数据进行对比。
所采用的计算方法例如可以包括统计, 具体来说, 可以根据事先确定的 标准将标记模块进行参数标记的取值范围划分为若干个区间, 然后统计出所 有标记参数落入各个区间的次数, 并进而得出其所占据的百分比。 在这一例 子中, 是标记参数在各个区间中的分布即为该物体表面图像所对应的特征因 数。 当然, 本领域技术人员很容易想到, 区间的划分可以根据试验结果而采 取不均匀的方式。
当然, 上述方法只是可行的计算方法中较为简单的一种, 具体实施时可 以采取更加复杂的计算方法, 以求获得更加精确的特征因数。 这在后面的说 明中也会进一步提到。
判断模块的作用如前文所述, 用于将计算获得的特征因数与结冰状况数 据库 203中的已有特征数据进行对比, 从而找到与当前特征因数最接近的特 征数据。 该特征数据所对应的结冰状况(包括结冰种类、 结冰厚度和 /或结冰 面积等)即可认为是物体表面的当前结冰状况。 当然, 在判断的过程中还可 以引入新的参考量, 例如由温度传感器 111获得的环境温度。
结冰状况数据库 203是通过大量模拟试验以及对实际探测结果进行处理 和分类而获得的。 其中可以包括若干条数据, 每条数据都包括一种特定结冰 状况的信息(结冰种类、 结冰厚度和 /或结冰面积等)以及与该结冰状况对应 的特征数据 , 以供判断模块将特征数据与计算获得的特征因数进行对比而得 到所对应的结冰状况。
为了更好地理解本发明的内容, 下面将就本发明的工作原理进行简单的 说明。
无论是在可见光、 红外波段还是紫外波段内, 冰层的光学特性(冰层一 空气界面反射、 冰层内的散射以及吸收等)都随着结水状况的改变而呈现出 不同, 从而形成差异明显的结冰图像。 结冰与没有结冰的图像, 不同种类不 同厚度结冰之间的图像的差别是非常明显的。
在考察结冰种类时, 可以考察图像的特性例如各个像素亮度值(包括灰 度亮度和三基色亮度)的均匀性。 当结冰为明冰时, 由于冰层内部近似于透 明, 因此冰层和空气界面上反射的电磁波能够以较大的强度被传像光纤束所 接收, 因此图像像素的亮度值较大且均勾。 当结冰为凇冰时, 由于冰层内夹 杂空气泡, 反射效应大为降低, 散射效应较强, 因此图像像素亮度值较低且 分布不均。 而混合型冰则介于上述两种情况之间。
在考察结冰厚度时, 也同样可以考察冰层的亮度。 因为, 当确定了结冰 类型以后, 在一定结冰厚度范围内, 结冰厚度愈厚, 则图像像素点的亮度也 愈大。
下面将通过两个例子来对结水分析单元 202的整个工作过程进行描述, 以便更加清楚地了解其工作原理及优点。
在第一个例子中, 在接收到图像固定装置传输的图像时, 结冰分析单元 202中的取点模块首先根据事先确定的规则从上选取若干个像素点 (例如 N 个)。 所谓的规则是指取点可以只在图像的特定区域中进行, 也可以在某些 区域选取得相对密集而在其它区域选取得相对稀疏等等。 然后, 由参数标记 模块通过图像处理领域中的常用软件对每个选取的像素点进行三基色分析, 分别得到每个像素的三基色值, 从而完成参数标记。 以八位的微处理器系统 为例, 各个基色的取值范围均在 0-255之间。
标记完成的三基色值被传输至计算模块。 首先根据事先划分好的三基色 取值区间将每个点归入其对应的区间中。每种三基色值的划分可以根据需要 进行, 例如将红光、 绿光和蓝光分别均勾或不均勾地划分为 p、 q和 r个段, 从而一共形成了 K=p x q x r个三基色值区间。然后统计出落入各个区间的点 的数目 n〗、 n2、 n3…… ηκ以及占总点数 N的百分比 m2、 m3…… mK。 该 点的数目 {nh n2, n3 ... ηκ}或者百分比 {xh x2, x3 ... χκ}即用作与当前表面图 像相对应的特征因数。
最后通过判断模块将计算获得的特征因数与结冰状况数据库 203中存有 的特征数据进行对比, 从而选定在数据库中与目前的结冰状况最接近的一条 数据, 并以该条数据中所包含的结冰状况的信息 (结冰种类、 结冰厚度和 / 或结冰面积等)作为物体表面的当前结冰状况。
而在第二个例子中, 参数标记模块和判断模块的工作过程并无太大区 另' L 但是在计算模块中却采用了不同的计算方法。
在考察结冰种类时, 可釆用像素亮度值的方差来作为判断的特征因数, 这样可以更加明显地看出亮度的分布; 而在考察冰层厚度时, 可将所有像素 点亮度值的总和作为最终的特征因数。 通过将计算获得的所有像素点亮度值 的方差和总和与结冰状况数据库 203中的已有特征数据进行对比, 即可方便 地得到结冰厚度大小的定量探测结果。
此外, 虽然没有仔细说明, 但是本领域技术人员可以想到, 上述探测器 和探测方法也可以实现对侧面观察到的冰层厚度的直接识别求解。 如图 5所 示, 通过对图像中灰度和颜色等进行识别, 可以清楚地区分出结冰区域和物 体表面区域, 并且通过对结冰图像进行分析, 选取多个测量点求出其冰层厚 度的平均值, 也可以很容易地求出整个冰层的厚度。
除了上述实施例以外, 本领域技术人员还可以想到其它改进措施, 以进 一步提高性能。 例如, 可以在获得各个点的灰度和 /或色谱值后, 不是直接以 其进行计算, 而是将其与清洁无冰图像中的灰度和 /或色傳值进行对比后, 以 其差值作为后续计算判断的标记参数; 还可以将物体表面分为若干个区域, 然后针对每一个区域都分别独立计算判断, 以使各区域的判断结果相互印 证, 从而降低探测的误差。
下面将对除了图像处理系统 2-Α以外的功能单元进行筒单描述, 其中, 这些功能单元均可采用现有技术中已有的方案, 不属于本发明的内容。
温度测控系统 2-Β能够获取温度传感器 111的温度信号, 并将温度信号 与设定的温度值相比较, 从而作为加热器 112工作的控制基础。 并且也可以 将温度值传输给判断模块, 作为判断当前结冰状况的参考量。
光源控制单元 2- C控制电磁波发射装置 1 15的工作, 即对电磁波种类、 发射时间和发射功率等进行控制。 电磁波发射装置 115可以连续工作, 也可 以规律或不规律地间断工作。 在规律的间断工作状态下, 电磁波的发射频率 可以选为 1-20 Hz, 这既可和图像获取系统协调工作, 又能保证有足够快的 探测速度。
在非连续的工作状态下, 可以通过中央微处理器 2-D对光源控制单元 2-C和图像处理系统 2-A的工作进行协调, 使得仅在电磁波发射装置 115工 作时, 图像处理系统 2-A才进行工作。
中央微处理器 2-D可以实现对图像处理系统 2-A、 温度测控系统 2- B以 及光源控制单元 2-C的控制和相互之间的信息交换,从而实现各单元的功能。
下面将参照图 3和 4, 以飞行器的结水探测为例, 说明根据本发明的结 冰探测器及探测方法的两种应用方式。 其中, 探测器的内部结构与前一实施 方式的结构大致相同, 因此不再重复。
根据本发明的结水探测器及探测方法可以用于对物体表面较小区域的 结水状况进行微观探测。 例如图 3所示, 图像获取系统的前端 110埋在飞行 器表面中, 并且朝向外侧, 传像光纤束在保护套 106中从前端 110引出并延 伸至远离该前端的图像处理系统(图中未显示) 。 在该应用中, 聚焦镜头选 用的是微距镜头, 所能获取的图像仅限于前端所对准的非常有限的面积。 但 是由于距离非常近, 因此可以实现对冰层内部图像的 观探测, 所获取的图 像信息的精确度非常高。 随之求得的单位时间内结冰厚度的增加量即结冰速 率也相应地更加精确。
需要补充说明的是, 在此布置方式下, 探测器前端结冰不可避免, 否则 将无法实现结冰探测。 但是此时, 仍然可以设置除冰装置, 以用于探测器的 复位。 的结冰状况进行宏观探测。 如图 4所示, 探测器安装于飞行器的垂尾上, 并 斜对着尾翼表面。 在该应用中, 聚焦镜头选用长焦镜头或是鱼眼广角镜头, 并且通过调整镜头的参数, 可以实现对整个需要进行结冰探测的表面区域 (例如, 图中所示为 a-b-c-d的矩形区域) 均获取清晰的图像。 在此布置方式下, 探测器前端不能存在结冰, 否则将无法获取待探测的 物体表面的结冰图像。 此时, 防冰除冰装置的设置就显得非常重要。 并且由 于传像光纤束是耐高温的玻璃光纤或石英光纤, 故只要加热装置的温度不是 太高, 就不会损坏探测器。
这种针对物体表面较大区域的探测形式虽然在对局部点进行探测上精 度较前一种形式低, 但是可以实现对整个区域范围的整体探测。 从这一角度 来说, 又能提高对结冰状况的总体分析精度。 因为冰层分布的不均勾可能造 成点探测的结果并不能代表整体的结冰状况, 使得依据几个特定点得出的结 论偏离实际情况。
除此以外, 这种宏观探测形式在某些特殊的应用中还能产生特殊的技术 效果, 例如实现对由过冷大水滴形成的 "后流冰" 的探测。 这对于飞行器等 领域的结冰探测来说, 具有非常重要的意义。
所谓的过冷大水滴是指中位容积直径范围超过 50微米的过冷水滴。 由 于过冷大水滴具有较大的质量, 因此在结水之前需要放出大量的潜热。 其在 接触例如飞行器表面后的一段时间内仍然保持液体状态, 不会发生结冰, 只 有当液体的潜热完全释放出来后, 结冰才会发生在沿气流方向向后一定距离 的表面上。 因此, 在 "后流冰" 的情况下, 会出现在例如飞行器机翼和尾翼 的前缘部位没有结冰, 而在前缘之后的非防护部位发生结冰的特殊情况。
根据传统的探测方法, 如果要探测这种结冰, 需要在较大的部位上设置 很多结冰探测器单元。 这样, 不仅要求有大的安装空间, 还会对物体表面的 结构造成破坏, 并且设置多个结冰探测器单元也会显著增加成本。 则可以轻易地实现对 "后流水" 的探测。 所要做的仅仅是修改计算模块中算 法 (例如将待探测的物体表面沿气流方向分区, 并对每一区域分别计算) , 使之能够实现对物体表面上沿着气流方向前一段没有结冰, 而后一段结冰的 情况进行识别, 就可以实现后流冰探测。
以上描述的是本发明的优选实施方式。 但是应当理解的是, 本领域技术 人员在阅读了上述说明后, 能够很容易想到其它实现本发明的具体方式, 而 这些具体方式是显而易见的。发明人预期本领域技术人员可以实施合适的改 变, 并且这些变化都应当被包括在由权利要求书所限定的保护范围中。

Claims

权 利 要 求 书
1. 一种图像结水探测器, 包括: 图像获取系统 (1-A )和图像处理系统 ( 2-A ) , 其中,
所述图像获取系统(1-A ) 能够获取物体表面的图像,
所述图像处理系统 (2-A ) 能够对所述图像进行分析, 从而得到所述物 体表面的结冰状况。
2. 如权利要求 1所述的图像结冰探测器, 其特征在于, 所述图像处理系 统(2-A )主要包括结冰分析单元(202 ) , 其包括标记模块、 计算模块和判 断模块, 其中, 所述标记模块用于对所述图像标记与结冰状况相关的若干参 数, 所述计算模块用于对所标记的参数进行计算, 以获得所述图像的特征因 数, 所述判断模块用于根据所述特征因数判断得到所述物体表面的结冰状 况。
3. 如权利要求 1或 2所述的图像结冰探测器, 其特征在于, 所述标记模 块进行标记的依据是所述图像的亮度。
4. 如上述任一项权利要求所述的图像结冰探测器, 其特征在于, 所述标 记模块包括灰度分析模块和 /或色谱分析模块,分别用于通过灰度或色谱对所 述图像标记若干参数。
5. 如上述任一项权利要求所述的图像结冰探测器, 其特征在于, 所述图 像处理系统 (2-A )还包括取点模块, 用于从所述图像上获取至少一部分像 素点, 并由所述标记模块对这些像素点标记参数。
6. 如上述任一项权利要求所述的图像结冰探测器, 其特征在于, 所述计 算模块计算所述特征因数的依据是所标记参数的大小和 /或分布。
7. 如上述任一项权利要求所述的图像结冰探测器, 其特征在于, 所述图 像被划分为多个区域, 所述计算模块针对各个区域分别计算所述特征因数。
8。 如上述任一项权利要求所述的图像结冰探测器, 其特征在于, 所述计 算模块通过统计的方法获得所述特征因数。
9. 如权利要求 8所述的图像结冰探测器, 其特征在于, 所述标记模块进 行参数标记的范围被划分为多个区间, 所述计算模块将所标记的参数在所述 多个区间内的分布作为所述特征因数。
10. 如权利要求 8所述的图像结冰探测器, 其特征在于, 所述计算模块 计算所标记参数的方差和 /或总和, 并将其作为所述特征因数。
11. 如上述任一项权利要求所述的图像结冰探测器, 其特征在于, 所述 图像处理系统 (2-A )还包括结水状况数据库 (203 ) 。
12. 如权利要求 11所述的图像结水探测器, 其特征在于, 所述结冰状况 数据库 ( 203 ) 包括与各种结冰状况相对应的特征数据, 用于和所述特征因 数进行对比。
13. 如上述任一项权利要求所述的图像结冰探测器, 其特征在于, 所述 结冰状况包括结)水种类、 结冰厚度和 /或结冰面积。
14. 如上述任一项权利要求所述的图像结冰探测器, 其特征在于, 所述 单元 (201 ) 。
15. 如权利要求 14所述的图像结冰探测器, 其特征在于, 所述结冰状况 数据库 (203 ) 包括所述物体表面的清洁图像。
16. 一种飞行器结冰探测器, 其特征在于, 包括如上述任一项权利要求 所述的图像结冰探测器。
17. 如权利要求 16所述的飞行器结冰探测器, 其特征在于, 所述图像获 取系统 (1- A ) 的前端靠近待探测的物体表面而设置, 用于对所述物体表面 进行近距离的微观探测。
18. 如权利要求 16所述的飞行器结冰探测器, 其特征在于, 所述图像获 取系统 (1-A ) 的前端远离待探测的物体表面而设置, 用于对所述物体表面 进行远距离的宏观探测。
19. 一种探测物体表面结冰状况的方法, 包括以下步骤:
- 获取待探测的物体表面的图像,
-对所述图像进行分析, 从而得到所述物体表面的结冰状况。
20. 如权利要求 19所述的方法,其特征在于,对所述图像进行分析包括: - 给所述图像标记与结冰状况相关的若干参数,
-对所标记的参数进行计算, 从而获得所述图像的特征因数,
-根据所述特征因数得到所述物体表面的结冰状况。
21. 如权利要求 19或 20所述的方法, 其特征在于, 给所述图像标记参 数的依据包括图像的特征。
22. 如权利要求 21所述的方法,其特征在于,所述图像的特征包括亮度。
23. 如上述任一项权利要求所述的方法, 其特征在于, 所述参数标记通 过对图像的灰度和 /或色谱进行分析而进行。
24. 如上述任一项权利要求所述的方法, 其特征在于, 在给所述图像标 记参数之前, 先从所述图像上获取至少一部分待标记的像素点。
25. 如上述任一项权利要求所述的方法, 其特征在于, 给所述图像标记 参数包括将所述图像与所述物体表面未结冰时的清洁图像进行对比, 并将对 比后的结果作为标记的依据。
26. 如上述任一项权利要求所述的方法, 其特征在于, 对所标记的参数 进行计算的依据包括所标记的参数的大小和 /或分布。
27. 如上述任一项权利要求所述的方法, 其特征在于, 采用统计的方法 对所标记的参数进行计算。
28. 如权利要求 27所述的方法, 其特征在于, 将所标记参数的范围划分 为多个区间, 并将所标记的参数在所述多个区间内的分布作为所述特征因 数。
29. 如权利要求 28所述的方法, 其特征在于, 计算所标记参数的方差和 /或总和, 并将其作为所述特征因数。
30. 如上述任一项权利要求所述的方法, 其特征在于, 将所述图像划分 为多个区域, 并对各个区域分别判断其结冰情况。
31. 如上述任一项权利要求所述的方法, 其特征在于, 提供结冰状况数 据库 ( 203 ) , 用于和所述图像进行对比。
32. 如权利要求 31所述的方法,其特征在于,所述结冰状况数据库( 203 ) 包括与各种结冰状况相对应的特征数据, 通过将所述特征因数与所述特征数 据进行对比, 从而得到所述物体表面的结冰状况。
33. 如权利要求 32所述的方法, 其特征在于, 所述数据库包括关于结冰 种类、 结水厚度和 /或结冰面积的数据。
PCT/CN2011/075787 2010-07-02 2011-06-16 图像结冰探测器及探测方法 WO2012000384A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US13/807,964 US20130113926A1 (en) 2010-07-02 2011-06-16 Detecting device for detecting icing by image and detecting method thereof
EP11800129.6A EP2589928B1 (en) 2010-07-02 2011-06-16 Detecting device for detecting icing by image and detecting method thereof

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
CN 201010219377 CN102313512B (zh) 2010-07-02 2010-07-02 图像结冰探测方法
CN201010219377.8 2010-07-02
CN 201010219357 CN102313510B (zh) 2010-07-02 2010-07-02 图像结冰探测器
CN201010219357.0 2010-07-02

Publications (1)

Publication Number Publication Date
WO2012000384A1 true WO2012000384A1 (zh) 2012-01-05

Family

ID=45401384

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/075787 WO2012000384A1 (zh) 2010-07-02 2011-06-16 图像结冰探测器及探测方法

Country Status (3)

Country Link
US (1) US20130113926A1 (zh)
EP (1) EP2589928B1 (zh)
WO (1) WO2012000384A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113636086A (zh) * 2021-10-13 2021-11-12 中国空气动力研究与发展中心低速空气动力研究所 一种基于导光板的结冰探测装置及结冰探测方法
CN115214894A (zh) * 2022-08-31 2022-10-21 中国空气动力研究与发展中心低速空气动力研究所 一种精准确定结冰传感器最优安装位置的方法和存储介质

Families Citing this family (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8907798B2 (en) * 2012-01-05 2014-12-09 The Boeing Company Supercooled large drop icing condition detection system
US9180972B2 (en) 2012-01-05 2015-11-10 The Boeing Company Supercooled large drop icing condition detection system
CN103101626B (zh) 2012-12-04 2015-09-02 中国商用飞机有限责任公司 结冰探测器
CN102967268B (zh) * 2012-12-06 2015-12-02 国网重庆市电力公司长寿供电分公司 测定架空输电线路覆冰厚度的便携式覆冰观测仪及其方法
US10689122B2 (en) 2016-03-29 2020-06-23 Sikorsky Aircraft Corporation Icing detection systems
US10146049B2 (en) * 2016-04-01 2018-12-04 Flir Systems, Inc. Cold weather camera kit
US10124900B2 (en) 2016-06-28 2018-11-13 Rosemount Aerospace Inc. Automated super-cooled water-droplet size differentiation using aircraft accretion patterns
US10214299B2 (en) 2016-09-22 2019-02-26 The Boeing Company Light detection and ranging (LIDAR) ice detection
US10737792B2 (en) 2016-09-22 2020-08-11 The Boeing Company Turbofan engine fluid ice protection delivery system
US10252808B2 (en) 2016-09-22 2019-04-09 The Boeing Company Fluid ice protection system flow conductivity sensor
US10429511B2 (en) * 2017-05-04 2019-10-01 The Boeing Company Light detection and ranging (LIDAR) ice detection system
US10611488B2 (en) 2017-05-05 2020-04-07 Rosemount Aerospace Inc. Ice accretion boundary locator
US10450075B2 (en) * 2017-08-28 2019-10-22 Rosemount Aerospace Inc. Method of making a magnetostrictive oscillator ice rate sensor probe
US11125157B2 (en) 2017-09-22 2021-09-21 The Boeing Company Advanced inlet design
FR3074145B1 (fr) * 2017-11-28 2022-04-29 Zodiac Aerotechnics Procede de detection de la presence de cristaux de glace dans un environnement
DE102017130593A1 (de) * 2017-12-19 2019-06-19 Airbus Defence and Space GmbH Quantitative Blockierungsanalyse eines Gitters in Vereisungsbedingungen
US10160550B1 (en) * 2017-12-21 2018-12-25 United Technologies Corporation Ice detection and real time ice estimation using innovative sensor data learning approaches
US10669035B2 (en) * 2018-03-02 2020-06-02 The Boeing Company Detecting the presence of ice on a surface using ultraviolet light
BR102022003769A2 (pt) 2021-05-04 2022-11-16 The Boeing Company Estrutura de entrada de nacele de um conjunto de motor
FR3123426B1 (fr) * 2021-05-25 2024-02-09 Airbus Operations Sas Dispositif de mesure de l’epaisseur d’une accretion de givre sur une surface et aeronef muni d’un tel dispositif
CN113719433A (zh) * 2021-06-15 2021-11-30 国家电投集团宁夏能源铝业中卫新能源有限公司 风电机组叶片结冰检测方法、装置、系统及存储介质
CN113625606B (zh) * 2021-08-11 2022-11-15 上海秉越电子仪器有限公司 一种医用无尖刺冰泥制作控制系统
CN114194400B (zh) * 2022-02-15 2022-04-29 中国空气动力研究与发展中心低速空气动力研究所 一种可自识别故障的光纤结冰探测装置及故障探测方法
CN115892477B (zh) * 2023-01-09 2023-05-02 中国空气动力研究与发展中心低速空气动力研究所 结冰范围的等离子体探测方法、装置及存储介质、程序
CN116624347B (zh) * 2023-05-19 2024-04-26 华润电力技术研究院有限公司 一种风机叶片结冰监测方法、系统及装置

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1996020465A1 (en) * 1994-12-28 1996-07-04 Kenneth Gustavsson System for zone surveillance
US5557261A (en) * 1994-05-06 1996-09-17 Nichols Research Corporation Ice monitoring and detection system
US5921501A (en) * 1993-07-14 1999-07-13 Northrop Grumman Corporation Aircraft ice detecting system
US5963148A (en) * 1995-03-23 1999-10-05 Honda Giken Kogyo Kabushiki Kaisha Road situation perceiving system
US20020158768A1 (en) 2001-04-25 2002-10-31 Severson John A. Inflight ice detector to distinguish supercooled large droplet (SLD) icing
US20040231410A1 (en) 2003-03-10 2004-11-25 Marc Bernard Large spectrum icing conditions detector for optimization of aircraft safety
CN101762239A (zh) * 2010-01-26 2010-06-30 中国气象局气象探测中心 电线积冰测量装置及测量方法

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB9027943D0 (en) * 1990-12-21 1991-02-13 Gec Ferranti Defence Syst Ice detector
US5180122A (en) * 1991-05-10 1993-01-19 Fmc Corporation Apparatus for deicing
US5450502A (en) * 1993-10-07 1995-09-12 Xerox Corporation Image-dependent luminance enhancement
JP3358260B2 (ja) * 1993-12-06 2002-12-16 株式会社ニコン 沈胴式ズームカメラ、及びこれに用いられる鏡筒装置
US5781115A (en) * 1995-04-03 1998-07-14 Target Microwave Apparatus and method for detection and thickness measurement of coatings over a surface
US7280134B1 (en) * 1998-01-26 2007-10-09 Thales Avionics, Inc. Landscape camera system with electronic field of view switching
US6166661A (en) * 1999-01-22 2000-12-26 Rockwell Collins, Inc. System for detecting ice on an aircraft
US6173231B1 (en) * 2000-01-31 2001-01-09 Navigation Technologies Corp. Method and system for collecting data concerning thermal properties of roads for a geographic database and use thereof in a vehicle safety system
US7586419B2 (en) * 2003-06-13 2009-09-08 Gkn Aerospace Services Limited Ice detection apparatus and method
WO2007135915A1 (ja) * 2006-05-23 2007-11-29 Kirin Techno-System Company, Limited 表面検査装置

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5921501A (en) * 1993-07-14 1999-07-13 Northrop Grumman Corporation Aircraft ice detecting system
US5557261A (en) * 1994-05-06 1996-09-17 Nichols Research Corporation Ice monitoring and detection system
WO1996020465A1 (en) * 1994-12-28 1996-07-04 Kenneth Gustavsson System for zone surveillance
US5963148A (en) * 1995-03-23 1999-10-05 Honda Giken Kogyo Kabushiki Kaisha Road situation perceiving system
US20020158768A1 (en) 2001-04-25 2002-10-31 Severson John A. Inflight ice detector to distinguish supercooled large droplet (SLD) icing
US20040231410A1 (en) 2003-03-10 2004-11-25 Marc Bernard Large spectrum icing conditions detector for optimization of aircraft safety
CN101762239A (zh) * 2010-01-26 2010-06-30 中国气象局气象探测中心 电线积冰测量装置及测量方法

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113636086A (zh) * 2021-10-13 2021-11-12 中国空气动力研究与发展中心低速空气动力研究所 一种基于导光板的结冰探测装置及结冰探测方法
CN115214894A (zh) * 2022-08-31 2022-10-21 中国空气动力研究与发展中心低速空气动力研究所 一种精准确定结冰传感器最优安装位置的方法和存储介质
CN115214894B (zh) * 2022-08-31 2022-11-22 中国空气动力研究与发展中心低速空气动力研究所 一种精准确定结冰传感器最优安装位置的方法和存储介质

Also Published As

Publication number Publication date
EP2589928A4 (en) 2014-11-05
US20130113926A1 (en) 2013-05-09
EP2589928B1 (en) 2020-10-14
EP2589928A1 (en) 2013-05-08

Similar Documents

Publication Publication Date Title
WO2012000384A1 (zh) 图像结冰探测器及探测方法
WO2012000385A1 (zh) 结冰探测器
CN102313510B (zh) 图像结冰探测器
CN105793691B (zh) 冰与水检测系统
CN102313512B (zh) 图像结冰探测方法
DK201200191U3 (da) System til detektering af ugunstige atmosfæriske tilstande foran et fly
US9769431B2 (en) Ice detector
CN103910062B (zh) 自动化的水滴测量和冰探测系统
CA2086508C (en) Apparatus for deicing
US20150120093A1 (en) Ice and water detection system
CN102253381B (zh) 用于机场跑道的异物自动检测系统及其检测方法
CN103043216B (zh) 结冰探测器
CN107728136A (zh) 一种机场跑道异物监测和清除引导系统及方法
CN106781230A (zh) 基于红外光谱技术的公路结冰检测与预警系统及其方法
EP3501990A1 (en) Quantitative blockage analysis of a grid in icing conditions
CN111354152A (zh) 一种火焰检测与定位系统
CN110097531A (zh) 一种用于无人机电力巡检的异构图像融合检测方法
JP2007316049A (ja) 路面モニタリングシステム
CN110108655B (zh) 一种道路路面状况检测方法及检测系统
CN104777103A (zh) 一种视距能见度仪及其测量方法
CN103101627B (zh) 结冰探测器
CN103448913A (zh) 飞机除冰实时监测装置
CN104103183A (zh) 环境自适应交通管理参数自动修改方法及系统
CN116609024A (zh) 一种基于激光诱导荧光的结冰风洞3d冰形测量方法
CN104768809B (zh) 用于借助摄像装置探测窗玻璃上雨滴的照明装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 11800129

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2011800129

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 13807964

Country of ref document: US