TWI577965B - Detecting method of detecting status of mold - Google Patents

Description

Mold state detection method

The present invention relates to a detection method, and more particularly to a mold state detecting method.

A conventional mold manufacturing apparatus using an injection molding technique uses a mold including a male mold and a female mold. Specifically, when the model is to be manufactured, the manufacturing apparatus first moves the male mold toward the mother mold to perform mold clamping (ie, the male mold and the female mold are tightly coupled). Next, the manufacturing apparatus ejects the liquid raw material to the master mold and waits for it to be cooled and molded. After the cooling is completed, the manufacturing device moves the male mold away from the master mold, and uses the thimble to push the formed mold out of the male mold to complete the mold release.

However, in the prior art, the operator needs to manually check whether the state of the mold is abnormal after the mold is removed from the manufacturing device (for example, the thimble is not completely retracted, there is residue in the male mold or the mold is damaged), and the next manufacturing can be performed if there is no abnormality. . The aforementioned detection method requires labor and increases manufacturing costs.

In order to solve the above problems, a mold state detecting device has been proposed. The existing mold state detecting device pre-stores a set of sample images (that is, images taken at a specific viewing angle when the mold state is normal), and can capture the current image of the mold (ie, detect images) at the same viewing angle when the mold is demolded. Next, the mold state detecting device directly compares the sample image with the detected image to determine whether the mold state is abnormal.

However, due to errors in the mechanical operation (such as the position of each mold after the mold is not completely the same or the mold may be shaken due to vibration), the existing mold state detecting device does not consider the above error and does not perform the captured image. In the case of correction, the detection result will be out of alignment, and the accuracy will be low.

Therefore, the existing mold state detecting device has the above problems, and a solution to be more effective is proposed.

The main object of the present invention is to provide a mold state detecting method which can correct a detected image and perform detection using the corrected detected image.

In order to achieve the above object, the present invention provides a mold state detecting method comprising the steps of: a) capturing a detection image of a mold, and setting a detection area in the detection image; b) obtaining the same image of the mold. And setting the same area in the same position in the sample image, wherein the sample area has the same area size as the detection area; c) calculating an area displacement value according to the image in the detection area and the image in the sample area; d) correcting the detection image according to the displacement value of the region; and e) determining a detection result according to the sample image and the corrected detection image.

The invention can effectively improve the accuracy of the mold state detection by correcting the detected image.

1‧‧‧Mold condition detecting device

10‧‧‧Image capture module

12‧‧‧ memory

14‧‧‧Output module

16‧‧‧Warning module

18‧‧‧ bracket

20‧‧‧ Magnetic seat

200‧‧‧Magnetic shell

202‧‧‧magnetic components

204‧‧‧ permanent magnet

206‧‧‧ knob

22‧‧‧Lighting module

24‧‧‧Chassis

3‧‧‧ Processor

30‧‧‧Regional setting module

32‧‧‧Displacement calculation module

320‧‧‧Regional adjustment module

322‧‧‧Search Module

324‧‧‧Computation Module

326‧‧‧ Alignment module

328‧‧‧Pixel Displacement Calculation Module

34‧‧‧ Calibration Module

36‧‧‧Test module

360‧‧‧Differential image generation module

362‧‧‧Decision Module

364‧‧‧Statistical Module

38‧‧‧Detection module

40‧‧‧ translation module

50‧‧‧Mold

52‧‧‧Female model

54‧‧‧Male model

6, 6', 6'' ‧ ‧ test images

60‧‧‧Detection area

62‧‧‧Detection subregion

7, 7'‧‧‧ sample image

70, 72‧‧‧ sample area

S60-S68‧‧‧Mold state detection steps

S640-S644‧‧‧Regional displacement calculation steps

S680-S682‧‧‧ Test results decision steps

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a block diagram of a mold state detecting device according to a first embodiment of the present invention.

2 is a structural diagram of a displacement calculation module according to a first embodiment of the present invention.

FIG. 3 is a structural diagram of a detection module according to a first embodiment of the present invention.

Fig. 4 is a perspective view showing the appearance of a mold state detecting device according to a first embodiment of the present invention.

Fig. 5A is a schematic view showing the closing of a magnetic seat according to a first embodiment of the present invention.

FIG. 5B is a schematic view showing the opening of the magnetic seat according to the first embodiment of the present invention.

Fig. 6 is a flow chart showing a method of detecting a mold state according to a first embodiment of the present invention.

Fig. 7 is a partial flow chart showing a method of detecting a mold state according to a second embodiment of the present invention.

Figure 8 is a partial flow chart showing a method of detecting a mold state according to a third embodiment of the present invention.

FIG. 9 is a schematic diagram of a detected image according to a first embodiment of the present invention.

FIG. 10 is a schematic diagram of a sample image according to a first embodiment of the present invention.

FIG. 11 is a schematic diagram of an image of a detection area according to a first embodiment of the present invention.

Figure 12 is a schematic view showing an image of a sample area according to a first embodiment of the present invention.

FIG. 13 is a schematic diagram of a corrected detection image according to the first embodiment of the present invention.

DETAILED DESCRIPTION OF THE INVENTION A preferred embodiment of the present invention will be described in detail with reference to the drawings.

Referring to FIG. 1 to FIG. 3, FIG. 1 is a structural diagram of a mold state detecting device according to a first embodiment of the present invention, and FIG. 2 is a structural diagram of a displacement computing module according to a first embodiment of the present invention, and FIG. 3 is the first embodiment of the present invention. The detection module architecture diagram of the embodiment. The mold state detecting device 1 (hereinafter referred to as the detecting device 1) of the present invention mainly includes an image capturing module 10, a memory 12, and a processor 3 electrically connected to the above components.

The image capturing module 10 can capture the detected image of the mold (the mold 50 shown in FIG. 4) (the detected image 6 shown in FIG. 9) for subsequent detection. The memory 12 stores a sample image 7. Specifically, the sample image 7 refers to an image captured by the male mold 54 when the state of the mold 50 is normal. Preferably, the shadow The capture module 10 is an infrared camera, and the detection image 6 and the sample image 7 are infrared images. Thereby, the generated image is not affected by the illuminance change of the ambient light, and the image with stable brightness can be provided.

Continuing to refer to FIG. 9 to FIG. 13 , FIG. 9 is a schematic diagram of a detection image according to a first embodiment of the present invention, FIG. 10 is a schematic diagram of a sample image according to a first embodiment of the present invention, and FIG. 11 is a detection of the first embodiment of the present invention. FIG. 12 is a schematic diagram of an image of a sample area according to a first embodiment of the present invention, and FIG. 13 is a schematic diagram of a corrected image of the first embodiment of the present invention for illustrating how the present invention detects an image 6 Make corrections. In this example, the image size of the detection image 6 and the sample image 7 are both 8 pixels×8 pixels, but the detection image 6 and the sample image 7 may also be other image sizes, such as 7 pixels×7 pixels or 10 pixels×10. Pixels are not limited.

The processor 3 is configured to control the detecting device 1 and mainly includes a region setting module 30, a displacement calculating module 32, a calibration module 34, and a detecting module 36.

The area setting module 30 is configured to set the detection area 60 in the detection image 6 (the area size is 8 pixels×8 pixels as an example), and set the sample area 70 of the same area size at the same position in the sample image 7 (for example, the same 6 pixels × 6 pixels).

In another embodiment of the present invention, the processor 3 may further include a detection module 38 and a connection area setting module 30. The detecting module 38 is configured to detect whether the mold 50 is demolded, and trigger the image capturing module 10 to capture the detected image 6 after detecting the mold 50 being demolded.

The displacement calculation module 32 is connected to the area setting module 30 for calculating the area displacement value according to the image in the detection area 60 and the image in the sample area 70.

Preferably, the displacement calculation module 32 further includes an area adjustment module 320, a search module 322, and a calculation module 324. The area adjustment module 320 can reduce the sample area 70 (such as the sample area 72 shown in FIG. 12), and define the complex detection sub-area 62 in the detection area 60 (to clearly show the size of the area of the detection sub-area 62, FIG. 11 only indicates one The group detection sub-region 62), wherein each of the detection sub-regions 62 has the same region size as the reduced sample region 72 (for example, is also 4 pixels × 4 pixels).

For example, the area adjustment module 320 can shorten the four sides of the sample area 70 by one pixel, and obtain the reduced sample area 72 as shown in FIG. The area adjustment module 320 can define the detection area 60 as a detection sub-area 62 having the same area size as the reduced sample area 72. The number of detection sub-areas 62 is exemplified by nine, that is, the first detection sub-area. (including a region of pixels P ₁₁ -P ₁₄ , P ₂₁ -P ₂₄ , P ₃₁ -P ₃₄ , P ₄₁ -P ₄₄ ), a second detection sub-region (including pixels P ₁₂ -P ₁₅ , P ₂₂ -P ₂₅ , a region of P ₃₂ -P ₃₅ , P ₄₂ -P ₄₅ ), a third detection sub-region (including a region of pixels P ₁₃ -P ₁₆ , P ₂₃ -P ₂₆ , P ₃₃ -P ₃₆ , P ₄₃ -P ₄₆ ), a fourth detection sub-region (a region including pixels P ₂₁ -P ₂₄ , P ₃₁ -P ₃₄ , P ₄₁ -P ₄₄ , P ₅₁ -P ₅₄ ), a fifth detection sub-region (including pixels P ₂₂ -P ₂₅ , P) ₃₂ -P ₃₅ , P ₄₂ -P ₄₅ , P ₅₂ -P ₅₅ area), sixth detection sub-area (including pixels P ₂₃ -P ₂₆ , P ₃₃ -P ₃₆ , P ₄₃ -P ₄₆ , P ₅₃ -P a region of ₅₆ ), a seventh detection sub-region (a region including pixels P ₃₁ -P ₃₅ , P ₄₁ -P ₄₅ , P ₅₁ -P ₅₅ , P ₆₁ -P ₆₅ ), and an eighth detection sub-region (including pixel P ₃₂₎ -P ₃₅ , P ₄₂ -P ₄₅ , P ₅₂ -P ₅₅ , P ₆₂ -P ₆₅ area), ninth detection sub-area ( A region including pixels P _{33 -} P ₃₆ , P _{43 -} P ₄₆ , P _{53 -} P ₅₆ , P _{63 -} P ₆₆ ).

The search module 322 is connected to the area adjustment module 320 for searching for the detection sub-area 62 corresponding to the reduced sample area 72 in the complex detection sub-area 62. Preferably, the displacement calculation module 32 further includes a comparison module 326 connected to the search module 322 for comparing the pixel values of the complex detection pixels of the images in the detection sub-regions 62 with the reduced sample region 72. Whether the pixel values of the plurality of sample pixels of the image match, and then generate and send the comparison result to the search module 322.

Preferably, the displacement calculation module 32 compares the complex sample pixels of the image in the reduced sample region 72 with the complex detection pixels of the images in each detection sub-region 62 via the comparison module 326, and images the images. One of the detection sub-regions 62 having the smallest difference serves as the aforementioned detection sub-region 62.

The calculation module 324 is connected to the search module 322 for calculating the regional displacement value between the reduced sample region 72 and the matching detection sub-region 62 (taking the ninth detection sub-region as an example), wherein the regional displacement value includes the horizontal displacement value. And vertical displacement values. For example, the region displacement value between the reduced sample region 72 and the ninth detection sub-region may be (+1, +1), that is, the horizontal displacement value is +1 and the vertical displacement value is +1.

Preferably, the displacement calculation module 32 further includes a pixel displacement calculation module 328 coupled to the calculation module 324 for calculating the image of the complex sample pixels of the image in the reduced sample region 72 and the corresponding detection sub-region 72. The complex number detects the complex pixel displacement values between the pixels and passes to the calculation module 324 to cause the calculation module 324 to calculate the region displacement values based on the complex pixel displacement values. For example, the displacement calculation module 32 can calculate the first pixel displacement value (+1, +1) according to the pixels P ₃₃ and S _{22 , and} calculate the second pixel displacement value ( +1, +1 according to the pixels P ₃₄ and S ₂₃ ). Calculating a third pixel displacement value (+1, +1) according to the pixels P ₃₅ and S _24, calculating a fourth pixel displacement value (+1, +1) according to the pixels P ₃₆ and S ₂₅ , and so on, The calculated 16 sets of pixel displacement values are transmitted to the calculation module 324. Next, the calculation module 324 averages the received 16 sets of pixel displacement values to obtain a region displacement value (+1, +1). Although the average calculation is used in this example to calculate the regional displacement value, it is not limited thereto. In another embodiment of the present invention, the computing module 324 employs other computing methods (such as weighting operations, filtering the extreme values before performing the averaging operation or the intermediate value operation).

The correction module 34 is connected to the displacement calculation module 32 for correcting the detection image 6 according to the calculated region displacement value. Preferably, the processor 3 further includes a translation module 40 connected to the correction module 34 for panning and detecting the image 6 according to the regional displacement value. For example, as shown in FIG. 13, if the area displacement value is (+1, +1), the panning module 40 can move all the pixels of the detected image 6 to the horizontal forward direction (for example, to the right) by 1 pixel. And move 1 pixel toward the vertical direction (for example, downward) to obtain the corrected detection image 6' (as shown in FIG. 13).

The detection module 36 is connected to the calibration module 34 for determining the detection result based on the sample image 7 and the corrected detection image 6'. Specifically, the detection module 36 may perform pre-detection processing (such as high-pass filtering processing, cropping processing, or brightness adjustment processing) on the sample image 7 and the corrected detection image 6′, respectively, and then compare the processed images. To improve the accuracy of detection.

Taking the cutting process as an example, the detecting module 36 can calculate the cutting value (such as the maximum value of the horizontal displacement value and the vertical displacement value) according to the regional displacement value or directly obtain the preset cutting value (such as 1 image). And according to the cutting value, the corrected detection image 6' and the sample image 7 (for example, cutting out the outer 1 pixel image) are respectively cut to obtain the processed detection image 6'' and the processed sample image respectively. 7', wherein the processed detection image 6'' and the processed sample image 7' have the same image size.

It is worth mentioning that the cutting process of the detection image 6 of the present invention is mainly for providing a translation space for performing the aforementioned translation. Specifically, if the cropping process cuts out the image of the outer 1 pixel, the detected image 6 can be shifted by up to 1 pixel up or down, and/or 1 pixel to the left or right, if the cropping process is cropped. When the image of the outer 2 pixels is removed, the detected image 6 can be shifted up to 2 pixels up or down, and/or 2 pixels to the left or right, and so on.

In another embodiment of the present invention, the detection module 36 includes a difference image generation module 360 and a decision module 362. The difference image generation module 360 is configured to subtract the processed sample image 7' from the processed detected image 6'' to obtain a difference image. The decision module 362 is connected to the difference image generation module 360 for determining the detection result according to the difference image.

Preferably, the detection module 36 further includes a statistics module 364 connected to the decision module 362. The statistics module 364 is configured to count the number of pixels in the difference image whose pixel value is greater than the preset threshold pixel value, and transmit the counted number of pixels to the decision module 362. Next, the decision module 362 determines whether the number of pixels is greater than a critical quantity, and if so, determines that the detection result is abnormal, otherwise the determination detection result is normal.

In another embodiment of the present invention, the detecting device 1 further includes an output module 14 (such as a display, a speaker or a printer) electrically connected to the processor 3 for outputting the detection result.

In another embodiment of the present invention, the detecting device 1 further includes an alert module 16 (such as a buzzer, an alarm or a warning light) electrically connected to the processor 3 for issuing an alert. Specifically, when the processor 3 determines that the detection result is abnormal, the processor 3 can control the warning module 16 to issue an alert.

Continuing to refer to FIG. 4, it is a schematic view of the appearance of a mold state detecting device according to a first embodiment of the present invention. In the present embodiment, the mold 50 includes a movable male mold 54 and a stationary female mold 52. Moreover, the processor 3, the memory 12, the output module 14, and the warning module 16 are disposed in the casing 24. Image撷 The module 10 is fixed to the master mold 52 via the bracket 18 to image the male mold 54. The detecting device 1 further includes a lighting module 22 (such as an infrared lamp) for illuminating the male mold 54.

Preferably, the bracket 18 further includes at least one magnetic seat 20. The magnetic seat 20 can be attracted to the magnetically conductive outer casing (such as an iron casing) of the mold 50 via a magnetic force.

Continuing to refer to FIG. 5A and FIG. 5B, FIG. 5A is a schematic view showing the magnetic seat closing according to the first embodiment of the present invention, and FIG. 5B is a schematic view showing the opening of the magnetic seat according to the first embodiment of the present invention.

The magnetic base 20 includes a magnetic conductive housing 200 (such as a steel housing), at least one magnetic isolation component 202 (such as a copper component) disposed in the magnetic conductive housing 200, and is movably disposed in the magnetic conductive housing 200. The permanent magnet 204 and the knob 206 of the permanent magnet 204 are linked.

As shown in FIG. 5A, when the knob 206 is rotated to a specific angle, the permanent magnet 204 can be rotated to the closed position, that is, both poles of the permanent magnet 204 are located adjacent to the magnetic isolation element 202. Since the magnetic lines of force have the characteristics of the shortest path from the N pole to the S pole, the two poles of the permanent magnet 204 at this time form a closed magnetic line of force inside the magnetic conductive housing 200, and cannot be adsorbed to the external magnetic conductor (ie, the guide of the mold 50). Magnetic enclosure).

As shown in FIG. 5B, when the knob 206 is rotated to another angle, the permanent magnet 204 can be rotated to the open position, that is, the magnetic isolation element 202 is adjacent to the position where the two ends of the permanent magnet 204 meet. At this time, the permanent magnet 204 is shielded by the magnetic isolation element 202 due to the shortest path of the magnetic lines of force, and a closed magnetic line of force cannot be formed inside the magnetic conductive case 200, and the open magnetic line of force is formed by the magnetic conductive case 200 and the external magnetic conductive material. Since the magnetic conductive housing 200 and the external magnetic conductor are located on the magnetic flux path, the magnetic conductive housing 200 can be adsorbed on the external magnetic conductor.

It is worth mentioning that the foregoing area setting module 30, the displacement calculating module 32, the correcting module 34, the detecting module 36, the detecting module 38, the panning module 40, the area adjusting module 320, and the searching module 322 The calculation module 324, the comparison module 326, the pixel displacement calculation module 328, the difference image generation module 360, the decision module 362, and the statistics module 364 can be implemented by using a hardware module (such as an electronic circuit or burned). Record The integrated circuit of the bit circuit can also be implemented by a software module (such as a program or an Application Programming Interface (API)), but not limited thereto. When the aforementioned modules are implemented via a software module, the connection between the above modules refers to a link between programs.

When the module is implemented by a software module, the memory 12 can further store a computer program, and the computer program records a code or a machine code for implementing the foregoing modules. After the processor 3 executes the program code or mechanical code of the computer program, the functions of each module can be realized.

Continuing to refer to FIG. 6, a flow chart of a mold state detecting method according to a first embodiment of the present invention is shown. Specifically, after the processor 3 of the detecting device 1 executes the aforementioned computer program, the mold state detecting method (hereinafter referred to as a detecting method) of each embodiment of the present invention can be performed, and the detecting method includes the following steps.

Step S60: The detecting device 1 captures the detected image 6 of the male mold 54 via the image capturing module 10, and sets the detection area 60 in the detected image 6. Preferably, the detecting device 1 captures the detected image 6 after the detecting mold 50 is demolded.

Step S62: The detecting device 1 reads the sample image 7 from the memory 12, and sets the sample region 70 at the same position in the sample image 7, wherein the sample region 70 and the detecting region 60 have the same region size.

Step S64: The detecting device 1 calculates the region displacement value according to the image in the detection region 60 and the image in the sample region 70.

Step S66: The detecting means 1 corrects the detected image 6 based on the calculated area shift value to obtain the corrected detected image 6'. Preferably, the detecting means pans the detected image 6 in accordance with the area shift value to obtain the corrected detected image 6'.

Step S68: The detecting device 1 determines the detection result based on the sample image 7 and the corrected detected image 6'. Preferably, the detecting device 1 performs pre-detection processing on the sample image 7 and the corrected detected image 6' to obtain the processed sample image 7' and the processed detected image 6'', and then The obtained image is processed to determine the detection result. Preferably, after determining the detection result, the detecting device 1 can further output the detection result via the output module 14.

Continuing to refer to FIG. 7, a partial flow chart of a mold state detecting method according to a second embodiment of the present invention is shown. The detection method of the embodiment differs from the detection method of the first embodiment in that the step S64 of the detection method of the embodiment further includes the following steps.

Step S640: The detecting device 1 reduces the sample region 70 in the sample image 7 to obtain the reduced sample region 72.

Step S642: The detecting device 1 defines a complex detecting sub-region 62 in the detecting region 60, and searches the complex detecting sub-region 62 for the detecting sub-region 62 in which the included image matches the image of the reduced sample region 72, wherein each detector The region 62 has the same region size as the reduced sample region 72.

Preferably, the detection device 1 determines the detection sub-region when the pixel value of the complex detection pixel of the image in any of the detection sub-regions 62 matches the pixel value of the complex sample pixel of the image in the reduced sample region 72. 62 coincides with the reduced sample area 72.

Preferably, the detecting device 1 compares the complex sample pixels of the image in the reduced sample region 72 with the complex detection pixels of the image in each of the detection sub-regions 62, and minimizes the image difference between the plurality of detection sub-regions 62. As the aforementioned matching sub-region 62.

Step S644: The detecting device 1 calculates the region displacement value between the sample region 72 and the matching detection sub-region 62.

Preferably, the detecting device 1 calculates a complex pixel displacement value between the complex sample pixel of the image in the sample region 72 and the complex detection pixel of the image in the matching detection sub-region 62, and calculates a value according to the complex pixel displacement value. Group area displacement value.

Figure 8 is a partial flow chart showing a method of detecting a mold state according to a third embodiment of the present invention. The detection method of the embodiment differs from the detection method of the first embodiment in that the step S68 of the detection method of the embodiment further includes the following steps.

Step S680: The detecting device 1 subtracts the sample image 7' and the detected image 6'' to obtain a difference image.

Preferably, the detecting device 1 performs pre-detection processing (such as high-pass filtering processing, cropping processing, or brightness adjustment processing) on the sample image 7' and the detected image 6', respectively, to obtain the processed sample image 7' and After processing the detected image 6'', the processed image is subtracted, or other comparison methods are applied to obtain the difference image.

Step S682: The detecting device 1 determines the detection result according to the difference image. Preferably, the detecting device 1 systematically counts the number of pixels in the difference image whose pixel value is greater than the critical pixel value. If the detecting device 1 determines that the number of pixels is greater than the critical number, it determines that the detection result is abnormal, and may further issue an alert via the warning module 16, otherwise, the determination result is normal.

The invention can effectively improve the accuracy of the mold state detection by correcting the detected image.

The above is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Therefore, equivalent changes to the scope of the present invention are included in the scope of the present invention. Bright.

S60-S68‧‧‧Mold state detection steps

Claims (9)

A mold state detecting method comprises the steps of: a) capturing a detection image of a mold, and setting a detection area in the detection image; b) obtaining the same image of the mold, and the same in the sample image The position is set to be the same area, wherein the sample area has the same area size as the detection area; c) calculating an area displacement value according to the image in the detection area and the image in the sample area; d) shifting the displacement value according to the area displacement value Detecting an image; and e) determining a detection result based on the sample image and the translated image after translation. The mold state detecting method according to claim 1, wherein the step a is: taking the detection image after the mold is demolded. The method of detecting a mold state according to claim 1, wherein the step c comprises the steps of: c1) reducing the sample area; c2) searching for a detection sub-area in which the image included includes an image of the sample area, wherein the detecting The area has a plurality of the detection sub-areas, and each of the detection sub-areas has the same area size as the reduced sample area; and c3) calculating the area displacement value between the sample area and the corresponding detection sub-area. The method of detecting a mold state according to claim 3, wherein the step c2 is when the pixel value of the complex detection pixel of the image in any of the detection sub-regions matches the pixel value of the plurality of sample pixels of the image in the sample region. And determining that the image of the detection sub-area matches the image of the sample area. The method of detecting a mold state according to claim 3, wherein the step c3 is respectively calculating a complex pixel displacement value between a complex sample pixel of the image of the sample region and a complex detection pixel of the image in the detection subregion, and according to the These pixel displacement values calculate the displacement value of the region. The mold state detecting method according to claim 1, wherein the step e comprises the following steps: E1) subtracting the sample image from the detected image to obtain a difference image; e2) determining the detection result according to the difference image. The method of detecting a mold state according to claim 6, wherein the step e2 calculates a number of pixels in the difference image whose pixel value is greater than a critical pixel value, and determines that the detection result is when the number of pixels is greater than a critical number. abnormal. The mold state detecting method according to claim 1, wherein the detection image and the sample image are infrared images. The mold state detecting method of claim 1, wherein the region displacement value comprises a horizontal displacement value and a vertical displacement value.