GB2275773A

GB2275773A - Video thermography for detecting subsurface flaws

Info

Publication number: GB2275773A
Application number: GB9403089A
Authority: GB
Inventors: Peter Neville
Original assignee: HORNE SUBSEA Ltd
Current assignee: HORNE SUBSEA Ltd
Priority date: 1993-02-18
Filing date: 1994-02-18
Publication date: 1994-09-07
Also published as: GB9303219D0; GB9403089D0

Abstract

A heat sensitive camera picks up an image of the surface of an object to be investigated; the thermal image is analyzed by computer to calculate the presence of subsurface flaws, and a three-dimensional display, representing the internal structure of the object is generated. The thermal image may be collected passively. The method may be used to investigate the wall thickness of a pipe wall from the exterior; in this case the computer may first calculate the thickness of the pipe wall from the thermal image to allow a three dimensional display of the interior of the pipe to be generated.

Description

DESCRIPTION SUB-SURFACE DETECTION AND MEASUREMENT The present invention relates to the detection and measurement of sub-surface defects (SSDs) and other internal features in physical objects such as pipes.

The National Non-Destructive Testing Centre (Atomic Energy Authority) Harwell Laboratory in Oxfordshire has devised a technique known as Thermal Transient Thermography, or PVT - Pulse Video Thermography. Experimentation has shown that subjecting a material (of uniform temperature) to a momentary steep temperature gradient across its thickness can reveal the presence of a sub-surface flaw. The rate at which the heat dissipates depends on the structure of the flaw, and the progress of the heat pulse inwards can be monitored by studying the change in pattern of the surface temperature. These patterns are captured with an infrared camera and recorded field by field on video tape.

The latter technique does have a number of problems and disadvantages associated with it. The equipment needed to generate the thermal pulse necessary for PVT is large and heavy, and not easily transportable in the field. The gas bottles required to cool the detector are likewise cumbersome and not particularly portable. Steady state temperatures in plant systems significantly above or below the ambient temperature require different methods of inducing transients. The Thermal Image processing / data handling software associated with this known technique is of limited functionality. Furthermore, transient generating devices are unsuitable for wide-ranging field conditions.

It is an object of the present invention to provide a method of sub-surface defect detection and measurement, based on a thermal image, which substantially overcomes or mitigates the problems and disadvantages associated with the above described known technique.

It has already been proven that the surface temperature of an object is altered by SSDs, so that they show up as differing colours in a thermal image.

This is due to the differing heat flow through differing thicknesses of an object. In the case of a pipe for example, the contents may corrode the pipe or lay deposits down, and the diffusion of heat at points where this occurs will differ form the norm. On the thermal image, this shows up as bands of colour.

While major changes in thickness of an object are easily visible, minor hue variations, representing areas of minute thickness-variation, may not be particularly easy to detect, and some small faults are not actually physically visible to the human eye.

The present invention relates to a method of detecting and measuring a sub-surface characteristics, such as a defect, in an object by collecting and analysing a thermal image thereof and producing a perspective, "three dimensional" display representative of the internal structure of the object.

The method allows the thermal image to be collected passively.

The present invention also provided an apparatus for an apparatus for detecting and measuring a subsurface defect in an object, comprising a thermal imaging camera, means for analysing the thermal image received from an object under test and generating a three-dimensional display image representative of the internal structure of that object.

The invention also comprises the use of a system comprising a thermal imaging camera interfaced to a computer for the passive detection of sub-surface defects in an object.

In contrast to the above cited prior art, in the present method and apparatus, thermal images may be collected passively, ie without making contact with or heating the object. It is not necessary (though it may, in some circumstances be desirable) to heat the object, since naturally occurring temperature variations can be used to establish the internal structure of an object.

The method is particularly well suited to detection of SSDs in pipes, since the internal surface of a pipe is typically at a substantially constant temperature (particularly if the pipe is filled with a fluid) which differs from the temperature of the outer surface of the pipe.

In the following description reference is made to the analysis of SSDs in pipes, but the invention may be used for analysis of SSDs in a wide variety of other objects, and indeed for detecting internal features other than defects.

A system suitable for carrying out the present method of SSDs detection comprises a thermal imaging device, such as an infra red camera, and an appropriately programmed computer with a graphics facility.

A preferred method of calculating pipe thickness from a thermal image produced by a system in accordance with the present invention is now described.

Accurate measurements of pipe thickness can be obtained from the thermal images produced by the present system based on the assumption that the internal pipe temperature is constant.

The varying levels of emmissivity for various types of pipe-coverings and metals can be ascertained, given the specific emmissivities and the thermal temperature of a particular pipe.

This is achieved by use of any of the three following formulae: 1) Thickness T = X - ((X/N-R)). (Z-R) (relative to a temperature range derived from the thermal imaging camera), where: T = the thickness at the measurement point X = the thickness of the pipe in the absence of defects.

N = (pixel value for ) the inside temperature of the pipe.

R =(pixel value for ) a reference temperature of a portion of the pipe whose thickness is equal to P.

Z = pixel value at the measurement point.

2) T = (Tmax - Tr). Tm T = Thickness at measurement point.

Tmax = Maximum temperature on I.T. Camera range.

Tr = Temperature at the reference point.

Tm = Temperature at measurement point.

Xr = Thickness at the reference point.

3) X2 = X1 + q K.A (T1-T2) X2T = Thickness at measurement point.

X1 = Thickness at reference point.

q = Rate of conductive heatflow of the material.

K = Thermal conductivity of the material.

A = Area of point being measured T1 = Temperature at reference point.

T2 = Temperature at measured point.

To calibrate the computer for a known reference point we use the formula: (Tr - Ti). (Z.Pi.K.L) q r = ri.X e Ti = Inside temperature of pipe.

Tr = Outside temperature of pipe.

r = Outside radius of pipe ri = Inside radius of pipe.

q = Rate of heat flow through the material.

K = Thermal conductivity of the material.

L = Length of pipe being measured.

It has now been discovered by the present applicants that the information gained from thermal imaging can be clarified by using three-dimensional graphics.

Using the present system, an area of interest chosen from the original thermal image may be presented in a perspective display. Areas of pipe thickness-variation are shown as troughs and peaks.

The height of the troughs and peaks may be determined using the above formulae, so that the display produced is a direct representation of the internal surface of the pipe. Very small defects, of too small a pixel grouping to be visible in the unprocessed thermal image, are visible in this format, as the three dimensional image is drawn pixel by pixel from the thermal image. The three dimensional display can be further clarified by placing an "isotherm limitation" on the display. This places limits on the information that may be displayed, such that only levels outside of chosen norms will be drawn.

The thermal image is analysed in the present system by a computer, which produces a three dimensional image of the area of interest. This is achieved by introducing a third value 'z', to the 'x' and 'y' positional co-ordinates of a point in the thermal image. 'z' is a colour value of the pixel at a particular 'x'/'y' position. The colour of this pixel is defined by the temperature recorded by the thermal image camera, and is one of a range of colours, e.g. 128 colours. A scale of colours to numbers, is established, such that higher/lower temperatures are given higher/lower numbers. To plot these values in perspective on the monitor, the 'y' and 'z' values are mathematically merged, and plotted against 'x'.A 'hidden line' function ensures that lines to the front of the display overwrite lower points at the rear of the display, so that the overall three dimensional illusion is retained.

The three dimensional display is far more explicit than the original thermal image, even though it contains no new information. It magnifies the chosen area by a factor of three, such that areas of potential interest are more easily visible. The "isotherm limitation" feature allows areas of potential interest to be isolated, which makes the end information obtainable by less skilled users. For users who have difficulties in interpreting thermal images and in translating thermal images into data, the three dimensional display is a simplified method of presenting the information for their perusal.

Conventional thermal imaging systems (TIS) come with their own monitor. However, since the software for the present system cannot be run on this dedicated screen, two monitors would be necessary when using a conventional TIS. By transferring the thermal image onto a normal PC monitor, as controlled by the present system, the TIS monitor can be dispensed with. In the present system, the screen may be divided, so that one area displays the perspective display and one area displays the thermal image. With the image right in front of the user, all the tools for its interpretation can be implemented easily and clearly (for instance, cursors and boxes can be superimposed on the image).

The conventional thermal image system has its own control keyboard, but this always has to be within a certain range of the thermal camera. Obviously there are major disadvantages with this, in that the software operator may not necessarily be anywhere near the camera. A feature of the present system is that it can be arranged to perform all the function of the TIS control keyboard remotely using a normal PC keyboard. This includes features such as changing the colour palette to make the thermal image more suited to the user's colour eye, changing the saturation and hues of the palette, and putting an isotherm limitation on the thermal image.

The present system can enable the user to zoom in to a chosen area of the image, with either x2, x4 or x8 magnification. Once a magnification level has been chosen, the user can pan around the entire thermal image generating corresponding perspective displays.

Zoom and pan may be performed on a real time image, or the image may be frozen and studied. The benefits of the zoom and pan features are that small defects visible to the eye in the original image can quickly be enlarged and studied before being discarded or taken even further (in 3D) as necessary.

Other facilities include the ability to perform any type of geometric calculation and measurement in the field of view; areas, volumes, distance, length of object sizes etc. can all be calculated. The constants needed to perform this task are the angle of view of the optical or imaging device and one known measurement in the field of view. The formulae vary from function to function. The accuracy is down to pixel level of the monitor.

Using the present system, the thermal image can be frozen, and an individual image can be written to file and saved. It can later be loaded back onto the screen, and treated as a usual frozen image, i.e.

subjected to zoom and pan, and have a three dimensional image created from it. The image can not only be loaded back into the system, but also into some other graphics packages, e.g. Paintbox in Windows.

The present system is non-contact and noninvasive. The heat of the pipe itself is used to generate the thermal image, and this method can be used whether the pipe is hotter or colder than the ambient temperature.

An entire system in accordance with the present invention can consist of a thermal imaging camera, its control box and a computer. For field-work, data can be stored onto video for later analysis, i.e. the only necessary gear to transport is the camera, control box and video. As the system is self-contained and portable, the number and types of places it is suitable for operation in is vastly increased compared to the conventional thermal imaging systems.

Claims

1. A method of detecting and measuring a sub surface defect in an object comprising collecting a thermal image thereof, analysing the thermal image and generating a "three dimensional" display image, representative of the internal structure of the object.

2. A method of detecting and measuring a sub surface defect in an object as claimed in claim 1, wherein the thermal image is collected passively.

3. A method of detecting and measuring a sub surface detect in an object as claimed in claim 1 or 2 wherein the thermal image is collected using a thermal imaging camera.

4. A method of detecting and measuring a sub surface defect in an object as claimed in claim 3, comprising mapping individual pixels of the thermal image onto the display.

5. A method as claimed in claim 5, wherein the "three dimensional" display image is achieved by introducing a third value (z) to the (x) and (y) positional co-ordinates of a point in the thermal image, the (z) value being a "colour" value of the pixel at any particular (x,y) position and the "colour" being defined by one of a plurality of possibly such "colours" recordable by the thermal image camera.

6. A method of detecting and measuring a sub surface in an object as claimed in any preceding claim, by calculating the internal structure of the object from the thermal image using the following formula: X = P~ P x (Z-R) where N-R X - the thickness of the object at the measurement point.

P = the thickness of the object in the absence of defects.

N = (pixel value for) an internal temperature of the object.

R = (Pixel value for) a reference temperature of a portion of the object whose thickness is equal to P.

Z = pixel value at the measurement point.

7. A method of detecting and measuring a sub surface defect in an object as claimed in any preceding claim, by calculating the internal structure of the object using the following formula: X = (Tmax - Tr) . Tm where Xr X = Thickness at measurement point.

Tmax = Maximum temperature on T.I Camera range.

Tm = Temperature at measured point.

Xr = Thickness at the reference point.

8. A method of detecting and measuring a sub surface defect in an object as claimed in any preceding claim by calculating the internal structure of the object using the following formula; X2 = X1 + (q) where K X A (T1-T2) X2 = Thickness at measurement point.

X1 = Thickness at reference point.

q = Rate of conductive heats low of the material.

K = Thermal conductivity of the material.

A = Area of point being measured.

T1 = Temperature at reference point.

T2 = Temperature at measured point.

9. A method of detecting and measuring a sub surface defect in an object as claimed in any preceding claim, by using the following formula for calibration: (Tr - Ti) x 2 (2 x Pi x K x L)/q r = ri x e Ti = Inside temperature of object.

Tr = Outside temperature of object.

r = Outside radius of object ri = Inside radius of object.

q = Rate of heat flow through the material.

K = Thermal conductivity of the material.

L = Length of pipe being measured.

10. A method of detecting and measuring a sub surface defect in an object as claimed in any of claims 6 to 9 by placing an isotherm limitation on the display so that only points of the thermal image whose measured temperatures are in one or more range (s) are mapped onto the display.

11. A method of detecting and measuring a sub surface defect in an object as claimed in any preceding claim by using a computer to analyse the thermal image and produce the display.

12. A method of sub surface defect detection substantially as herein described.

13. An apparatus for detecting and measuring a sub-surface defect in an object, comprising a thermal imaging camera, means for analysing the thermal image received from an object under test and generating a three-dimensional display image representative of the internal structure of that object.

14. An apparatus as claimed in claim 13, including means for establishing x and y co-ordinates for each point in the thermal image and means, for establishing z co-ordinate corresponding to each such point and based on the "colour" value of the pixel detected at that point established using a pre-stored colour scale.