JP2762321B2 - Thermal image measurement method and device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a temperature image measuring method and a temperature image measuring apparatus which can always perform high-precision temperature measurement by reducing a quantization error during temperature image measurement.

[0002]

2. Description of the Related Art Two-dimensional temperature images are widely applied to diagnosis of human bodies, defect inspection of electronic equipment, and failure diagnosis of thermal plants. A two-dimensional temperature image measuring device using a digital method measures a distribution of thermal energy radiated by an object as a thermal image using a CCD device or an image pickup tube, and converts the thermal image to an A-value.
This is a mechanism for converting into a digital temperature image by converting into a digital signal via a D converter (ADC). Taking an example of a temperature image measurement system using a CCD element, which has been remarkably developed in recent years, a thermal image is measured every 1/30 second with an element having approximately 500 × 500 pixels, and an 8-bit (256 divisions) The system converts the digital data into digital images by referring to the temperature conversion table.

However, when the temperature measurement range is set to 900 to 1600 ° C. and the A / D conversion is performed in 8 bits in the above temperature image measurement, the conversion is performed at 50 to 60 ° C. in a low temperature range.
Such a large conversion error occurs, and only a measurement result with a much lower accuracy than the temperature resolution of the CCD can be obtained. Such a conversion error is generally called a quantization error, and is regarded as a fundamental solution in digital image processing.

Normally, in order to improve the accuracy of AD conversion, means for increasing the resolution of the ADC is employed. For example, even if the same CCD is used, if a thermal image is measured and then converted to a digital image by a 12-bit ADC, the conversion section becomes 4096, and the temperature conversion accuracy is improved 16 times. However,
When the ADC is 12 bits, the image memory becomes 16 times larger and the operation circuit becomes complicated, so that the operation time becomes extremely long and the price of the image processing apparatus also becomes extremely high. For this reason, at present, this kind of high-precision image signal processing apparatus is not generally used except for a special measurement object.

[0005] In general image processing, a method of superimposing noise on an original signal to reduce a quantization error is widely known as a "dither method" for digitizing a television image. Techniques such as removing noise from an image signal using various spatial filters and extracting features are generally performed in image processing.

[0006]

As shown in FIG. 10, the dither method adds a dither signal (noise) created according to a certain rule to an input signal and binarizes the input signal with a constant threshold circuit. In this method, the error is reduced by using the method. On the other hand, the operation of the system for removing noise using a spatial filter is described as follows with reference to FIG. The center of the load matrix W (k, l) is a point in the input image f (m, n)
(i, j), the product of each element at the corresponding position in the overlapping partial image of f (m, n) is obtained, and the sum of them is calculated in the main image g (m, n). Value at point (i, j). This operation is performed at each position while shifting the position of the load matrix by one pixel.

However, no attempt has been made to apply these conventional dithering methods and spatial filtering methods to reduce quantization errors in temperature image measurement by combining them.

SUMMARY OF THE INVENTION An object of the present invention is to provide a temperature image measuring method and a device capable of always using a normal temperature image processing system by combining a dithering method and a spatial filtering technique and performing high-precision temperature measurement. Is to do.

[0009]

Means for Solving the Problems A temperature image measuring method according to the present invention for achieving the above object measures a thermal image as a thermal image in a thermal image processing system for converting a two-dimensional thermal image into a thermal image by a digital method. The obtained thermal image is converted to a thermal image by a temperature conversion table, and the generated thermal image is superimposed and averaged, or the thermal image is locally averaged. The configuration is characterized in that the quantization error during AD conversion is reduced by the conversion.

More specifically, in the above configuration, a normalization noise is superimposed on a detection signal measured by a CCD element or an image pickup tube, which is a thermal image detector, and A / D conversion is performed. Convert the temperature image to a temperature image, and superimpose the converted temperature image signal on the temperature image of multiple frames,
A preferred process of the present invention is to perform spatial filtering and averaging of the signal converted into the temperature image.

Table 1 shows a part of a typical temperature conversion table.

[0012]

[Table 1]

The first column of Table 1 is a bit order in which the luminance signal measured by the CCD element is 0 to 0 by an 8-bit AD converter.
Converted to a value of 255. The second column is the CCD output voltage value
(mV), which is the input signal to the AD converter. The last third column is the temperature conversion value (° C.). As can be seen from Table 1, for example, the temperature corresponding to 5.3 mV in the luminance signal of the first bit is 891 ° C., and 10.5% in the luminance signal of the second bit.
The temperature corresponding to mV is 955 ° C., the temperature in between cannot be identified. Therefore, a quantization error of up to 64 ° C. may occur in a low temperature range.

In the error reduction according to the present invention, as shown in the flow chart of FIG. 1, an AD conversion is performed after a noise n (t) is superimposed on an input signal s (t), and a luminance signal is converted into a temperature by a temperature conversion table. The output signal is converted and averaged to obtain an output signal θ (t). At this time, the noise applied to the input signal has a normal distribution of an average value x ₀ and a standard deviation value σ in amplitude p (n) as shown in the following equation (1). p (n) = 1 / (√ 2πσ) × exp (− (xx− ₀ ) ² / 2σ ² ) (1)

In the above equation (1), σ is normalized by the AD conversion step width as shown in the following equation (2). In the expression (2), Δs represents the step width of the AD conversion, and in the case of Table 1, it is about 5 mv. σ = (∫x ² × n (t) dt / ∫n (t) dt) / Δs ≒ 0.5 to 1.0 (2)

FIGS. 2 and 3 show changes in quantization error when noise is superimposed as described above. Among them, FIG. 2 shows a conversion curve of a temperature conversion value with respect to a luminance signal, which has a step-like pattern. FIG. 3 illustrates a temperature conversion state when normalized noise is superimposed in accordance with the process of FIG. 2 using this conversion table. In this case,
When the parameter σ exceeds 0.5, the output signal θ (t) becomes a smooth curve.

Table 2 shows an input voltage (corresponding to a luminance signal) corresponding to the target temperature and a temperature error when σ is changed. When σ is less than 0.5, the error increases, and when σ is more than 0.5, the error decreases. The reason why the error decreases as σ increases as described above is that bit values around the average value x ₀ of the input power are added and averaged. If σ is too large, the image quality naturally deteriorates. It is preferable to keep it within the step width. Therefore, the range of σ is 0.5 to 1.0 based on the step width of temperature conversion.
It is preferable to control within the range.

[0018]

[Table 2]

In the processing for averaging the temperature images, either a temporal averaging method or a spatial averaging method is used. Among them, an example of a flow of temporal averaging can be shown in FIG. In normal image measurement, images are generated by a two-dimensional scanning line method like a TV,
In FIG. 4, the scanning signal is the input signal s (t). A noise n (t) is superimposed on this, and after performing AD conversion and temperature conversion, an image is generated in the frame memory 1. Image is Δ
Generated at t intervals (1/30 second intervals on TV). This image is delayed by Δt, multiplied by a constant load coefficient w, and filed in another frame memory 2. When the operation for delaying by Δt is represented by z ⁻¹ , the following relational expression
The past image Is added in (3) is filed.

[0020]

The current image I ｛mΔt is added to the past added image Is
Adding｝ means time averaging.
The weighting factor is determined by the averaging time, with a small value being a long averaging time and a large value being a short averaging time. This method is not suitable for processing moving images. For example, when taking the time average of an image of 100 frames, 100 screens measured in 1/30 seconds must be added, so 3.3 seconds (100 × 1/30) is required. However, if the temporal change of the measurement object is slow, adaptation is possible.

The processing used for spatial averaging is spatial filtering as shown in FIG. If the noise generated in each image is not correlated, the temporal and spatial averages will be equal. An example of the flow of the local filtering is as shown in FIG. 5, but in this case, the selection of the load matrix is important. For the purpose of the present invention, filtering by a simple sum of products using a size of about 10 × 10 as the load matrix is appropriate. In the filtering, all pixels are sequentially multiplied by a matrix and added to calculate an average value.

FIG. 6 shows an example (11 × 11) of the load matrix. In this example, 1/4 of the load matrix is shown in a partially enlarged manner, but the other portions are the same. In addition, there is also an example in which the screen is divided into units of matrix and calculation is performed so that one representative temperature is assigned to one matrix. In this case, although the image quality is reduced, the pixels are greatly compressed and the calculation speed is reduced. There is an advantage of improving.

FIG. 7 is a block circuit diagram showing an example in which spatial filtering is realized by a digital signal processing circuit.
The circuit configurations from 2) to # 11 (i = 11) are the same as # 1 (i = 1). The input signal is delayed by a unit time (calculation of z ^-1 ) and multiplied by a load coefficient a _ij , and added, and the input signal is delayed and added by z ^-j to perform a calculation between scanning lines. In the case of this circuit, the product-sum operation of 11 × 11 pixels is 20 ms
Because it can be executed in ec, 1 /
It can be used for circuits that perform calculations for 30 seconds.

A temperature image measuring apparatus according to the present invention for use in the above-described temperature image measuring method includes a CCD camera 1 having a built-in thermal image detector, a camera controller and an amplifier 2 as shown in FIG. Circuit means for superimposing the normalized noise on the detection signal; an AD converter for converting a thermal image signal output from the circuit means into a temperature image by a temperature conversion table; An image processing apparatus 3 having a processing circuit means for averaging the whole or a spatial filter circuit means having a load matrix for spatial filtering and averaging a signal converted into a temperature image, and
The image processing apparatus 3 is connected to a monitor unit.

The CCD camera 1 is housed in a nozzle 4, and a filter 5 and a lens 6 are provided at a front stage thereof. As the thermal image detector incorporated in the CCD camera 1, a CCD element, an imaging tube, or the like is applied. CCD
The element usually has about 5 mV of noise and can be used as it is in many cases. However, depending on the conditions of the conversion table, it may be necessary to further reduce the noise level. For this reason, it is a preferable design to mount an electronic refrigeration element on the CCD element and cool it to reduce the noise level. In this case, for example, if the element cooling temperature is reduced to −20 ° C., the noise can be reduced to 2.5 mV.

According to the temperature image measuring method and apparatus according to the present invention, the first error reduction function that functions at the stage of AD conversion after superimposing normalized noise on the original signal, and averaging the converted temperature image The quantization error at the time of AD conversion can be significantly reduced by synergistic with the second error reduction function that functions at the stage of conversion. Therefore, a high-accuracy temperature image can always be measured, and accurate temperature measurement in a wide temperature range is guaranteed.

In particular, when spatial filtering is applied to the averaging process, not only a reduction in image quality due to superimposition of noise is suppressed, but also a secondary effect of eliminating fine noise on the screen. The resulting more accurate measurements can be made.

[0029]

EXAMPLE As shown in the overall view of FIG. 9, a temperature image measuring device 7 of the present invention was set on a furnace wall of a combustion furnace 8 via a nozzle tube 9. The device is connected to the monitor section of the instrument room 10 by a cable, and is controlled by a personal computer. The noise level of the CCD camera built in the device is adjusted to 5 mV, and the noise level corresponds to one bit of the temperature conversion table (Table 1). 11
A spatial filter circuit with a built-in x11 size load matrix is installed.

Depending on the apparatus and conditions, 1000 to 15
Table 3 shows the conversion temperature and error range when the combustion temperature at the 00 ° C. level was measured as the image temperature. In addition, the result of having measured the combustion temperature by the conventional temperature image measurement method is also shown in Table 3 as a comparative example. Comparing the results in Table 3,
In the comparative example, a quantization error at the time of AD conversion directly appears, and an error of about 20 to 50 ° C. in a low temperature part (at 1048 ° C.) and an error of about 3 to 5 ° C. in a middle temperature part (at 1251 ° C.). In contrast, in the case of the embodiment, it is recognized that the temperature is accurately measured in the entire temperature range. In addition, the image quality of the temperature image did not deteriorate due to the proper selection of the load matrix.

[0031]

[Table 3]

[0032]

As described above, according to the temperature image measuring method and apparatus according to the present invention, a wide range of temperature from a low temperature range to a high temperature range can always be measured accurately.
Therefore, it is extremely useful for temperature measurement operations in various industrial fields.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a flow of a temperature image measurement method according to the present invention.

FIG. 2 is a graph illustrating a relationship between a luminance signal and a temperature conversion value.

FIG. 3 is a graph illustrating a relationship between an input signal and an output signal.

FIG. 4 is a block diagram showing a flow of the present invention including a temporal averaging process.

FIG. 5 is a block diagram showing a flow of the present invention including a spatial averaging process.

FIG. 6 is an explanatory diagram exemplifying a load matrix by spatial filtering.

FIG. 7 is a block diagram showing a digital signal processing circuit for spatial filtering.

FIG. 8 is a schematic sectional view showing a temperature image measuring device of the present invention.

FIG. 9 is an overall explanatory diagram showing a measurement state according to an example.

FIG. 10 is an explanatory diagram of a dither method according to the related art.

FIG. 11 is an explanatory diagram of a spatial filtering method according to the related art.

[Explanation of symbols]

 Reference Signs List 1 CCD camera 2 Camera controller and amplifier 3 Image processing device 4 Nozzle 5 Filter 6 Lens 7 Temperature image measuring device 8 Combustion furnace 9 Nozzle tube 10 Instrument room

──────────────────────────────────────────────────の Continuing on the front page (72) Inventor Masahiro Sekiya 1 Kimitsu, Kimitsu City, Chiba Prefecture Nippon Steel Corporation Kimitsu Works (72) Inventor Yoshiyuki Shirakawa 1 Kimitsu, Kimitsu City, Chiba Prefecture Nippon Steel Corporation (56) References JP-A-62-265536 (JP, A) JP-A-62-193322 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) G01J 5/48 G01J 5/10 Practical file (PATOLIS) Patent file (PATOLIS)

Claims (3)

(57) [Claims] 1. A temperature image processing system for converting a two-dimensional thermal image in digital manner on the temperature image, thermal image
AD by superimposing normalized noise on the detection signal measured as
Conversion, the obtained thermal image is converted into a thermal image by a temperature conversion table , and the generated thermal image is superimposed and averaged or the thermal image is locally averaged to reduce the quantization error at the time of AD conversion. A temperature image measurement method characterized in that the temperature is reduced. 2. A method in which normalized noise is superimposed on a detection signal measured by a CCD element or an image pickup tube as a thermal image detector, and A
D conversion, the thermal image obtained by converting the temperature image by a temperature conversion table, or to the converted temperature image signal superimposing temperature images of a plurality of frames, a signal converted to a temperature image by spatial filtering average The method for measuring a temperature image according to claim 1, wherein the thermal image processing is performed. 3. A CCD camera having a built-in thermal image detector, a camera controller and an amplifier, circuit means for superimposing normalized noise on the detection signal, and a temperature conversion table for converting the thermal image signal output from the circuit means. the processing circuit means or the transformed signal to the temperature images for averaging the entire image by superimposing the temperature image to the AD converter and frame memory to convert temperature image space by filtering
A temperature image measuring device mainly comprising an image processing device incorporating a spatial filter circuit means incorporating a load matrix for performing averaging by weighting, and connecting the image processing device to a monitor section.