TWI639823B - Method, apparatus, and system for detecting thermal sealing defects - Google Patents

Abstract

A method, a device and a system for detecting a heat-seal defect. The device includes: a high frequency generator for applying a high frequency to a position of a metal film covered on a port of a container; and an infrared camera for applying the same to the container. Sensing infrared radiation from above to generate temperature data for each part of the container; a processing device for receiving the temperature data; converting the temperature data into an image; positioning the image to define an area of interest; calculation is not in The temperature within a qualified temperature range, the continuous area occupied in the region of interest of the image; the continuous area is calculated, and based on the calculation results, it is determined whether a seal defect is indicated; and an output device is used to indicate The seal is defective.

Description

Method, equipment and system for detecting heat seal defect

The present disclosure relates to a detection technology, and more particularly to a method, a device, and a system for detecting a heat seal defect.

In terms of detecting the thermal sealing defect of the container, the conventional technology generally uses an infrared sensor to sense the local temperature at the seal or the average temperature of the entire sealed area, and determine whether this temperature is within the normal temperature range, and judge based on this. Whether there are seal defects. Another conventional technique uses an infrared imager to obtain a thermal image of the container, and calculates the average temperature of the sealed defect area (such as a breach), compares it with a threshold value, and determines the sealing characteristics of the container accordingly. However, in the above-mentioned conventional technologies, because the temperature of the normal area is involved in the temperature calculation, the detection result is often inaccurate.

Therefore, it is necessary to propose a new detection technology to improve the lack of the conventional techniques mentioned above.

The purpose of the present disclosure is to provide a method, a device, and a system for detecting a heat seal defect, so as to improve the accuracy of the seal defect detection.

In order to achieve the above object, the present disclosure provides a method for detecting a heat seal defect, which is applied to detect a sealing characteristic of a port of a container, the port of the container is covered with a metal film, and a metal film is stacked on the metal film. The cover includes: applying a high frequency to the position of the metal film covered on the port of the container; using an infrared camera to sense infrared radiation above the container to generate temperature data of each part of the container; The temperature data of each part of the container is converted into an image of the container; the image of the container is positioned to define a region of interest (ROI); the temperature is not within a qualified temperature range, and the image is calculated in the image. The continuous area occupied by the region of interest; and calculating the continuous area, and determining whether a seal defect is indicated based on the calculation result.

Another aspect of the present disclosure is to provide a device for detecting a heat seal defect, which is used to detect the sealing characteristics of a port of a container, the port of the container is covered with a metal thin film, and a cover is stacked on the metal thin film. The device includes: a high frequency generator for applying a high frequency to the position of the metal film covered on the port of the container; and an infrared camera for sensing the high frequency after the high frequency generator applies the high frequency to the container. Infrared radiation to generate temperature data of each part of the container; a processing device coupled to the infrared camera to receive the temperature data generated by the infrared camera; convert the temperature data of each part of the container into the container Positioning an image of the container to define a region of interest (ROI); calculating a temperature that is not within a qualified temperature range and a continuous area occupied by the region of interest of the image ; Calculate the continuous area, According to the calculation result, it is judged whether a seal defect is indicated; and an output device is coupled with the processing device to indicate the seal defect.

Another aspect of the present disclosure is to provide a system for detecting a heat-sealing defect, which is used to detect the sealing characteristics of a port of a container, the port of the container is covered with a metal film, and a cover is stacked on the metal film. The system includes: a high-frequency generator is disposed on a production line, the production line has a conveyor belt for conveying a plurality of containers equipped with the metal film and the cover, and the high-frequency generator is used to correspond to each High frequency is applied to the position of the metal film covered on the port of each container; an infrared camera is used to sense infrared radiation above each container after the high frequency is applied by the high frequency generator to generate the parts of each container Temperature data; a processing device coupled to the infrared camera for receiving the temperature data generated by the infrared camera; converting the temperature data of each part of the container into an image of the container; Positioning to define a region of interest (ROI); calculate temperatures not within a qualified temperature range The continuous area occupied in the region of interest of the image; calculating the continuous area, and judging whether there is a sealing defect in the container according to the calculation result; and a removing device, coupled with the processing device, using The response processing device determines that the container has the sealing defect, and removes the container from the conveyor belt.

In the disclosed method, equipment and system for detecting heat seal defects, defect detection is performed for a region of interest, and a continuous area occupied by a temperature not within a qualified temperature range is calculated. Based on the calculation results, it is determined whether there is a seal defect . Since this disclosure uses a continuous area to track the temperature distribution of sealing defects, The analysis of the temperature data of the device is more accurate and can be closer to the expression form of the sealing defect. Therefore, compared with the conventional sealing defect detection method, a more accurate detection result can be obtained without reducing the detection speed.

11‧‧‧ container

12‧‧‧port

13‧‧‧Metal film

14‧‧‧ cover

15‧‧‧Insulation paper

20‧‧‧ System for detecting heat seal defects

21‧‧‧ conveyor belt

22‧‧‧High Frequency Generator

23‧‧‧ Infrared Camera

24‧‧‧Processing device

25‧‧‧ Display

26‧‧‧ Remove device

90‧‧‧ unit area

R1‧‧‧first radius

R2‧‧‧Second Radius

O‧‧‧ round

S31 ~ S37‧‧‧step

S41 ~ S47‧‧‧step

Figure 1 shows a schematic of a container.

FIG. 2 illustrates a system for detecting a heat seal defect according to an embodiment of the present disclosure.

FIG. 3 shows a method for detecting a heat seal defect according to an embodiment of the present disclosure.

FIG. 4 shows another method for detecting a heat seal defect according to an embodiment of the present disclosure.

Figures 5A and 5B show images without seal defects.

Figures 6A to 6C show images with sealing defects.

FIG. 7 is a schematic diagram showing a region of interest marked in an image of a container according to an embodiment of the present disclosure.

FIG. 8 is a schematic diagram showing another region of interest marked in an image of a container according to an embodiment of the present disclosure.

FIG. 9 is a schematic diagram of dividing a region of interest into a plurality of unit regions according to an embodiment of the present disclosure.

In order to make the objectives, technical solutions, and effects of the present disclosure more clear and definite, the following further describes the present disclosure in detail with reference to the drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present disclosure. The word "embodiment" is used as an example, example, or illustration, and is not intended to limit the present disclosure. In addition, the articles "a" and "an" used in this disclosure and the scope of the appended patents may generally be construed to mean "one or more" unless specified otherwise or clear from context to be singular.

The technique of detecting the heat-sealing defect in the present disclosure is applied to a heat-sealing situation in which a metal film generates heat with a high frequency. See Figure 1 for a schematic representation of a container. Generally, in the heat-sealing technology, before performing the heat-sealing action, the port 12 of the container 11 is covered with a metal film (generally an aluminum film) 13, and a heat-insulating paper 15 is attached to the inside of the cover 14 to cover the cover. 14 is screwed on from the port 12 of the container 11, and then the container 11 is sent to the production line. A high frequency is applied to the container 11. The high frequency causes eddy current on the metal thin film 13 and generates heat. This heat causes the The glue is melted so that the metal film 13 can adhere to the port 12 of the container 11.

Specifically, in a practical example, a lower surface of the metal thin film 13 is coated with a polymer film (not shown), and an upper surface thereof is coated with a layer of wax (not shown). When the cover 14 is screwed on, the heat-insulating paper 15 is attached to the metal film 13 through the extruded wax. When the metal film 13 is heated by high frequency waves, the wax on the metal film 13 melts, the metal film 13 is separated from the heat insulation paper 15 and moves to the port 12 of the container 11, and the polymer film below it becomes a fluid and flows to the port 12, Therefore, through the adhesive polymer film, the metal film 13 can seal the port 12.

Generally speaking, this method of sealing is used for medicines, health foods, or articles that require high sealing characteristics. This disclosure is not limited to the type of articles in containers. The type is not limited to the various configurations of the container, and any technology that uses a high frequency for heat sealing is within the scope of the present disclosure.

Please refer to FIG. 2, which shows a system 20 for detecting a heat sealing defect according to an embodiment of the present disclosure. The system 20 is disposed on a heat sealing production line of a container 11. As shown in FIG. 2, the production line has a conveyor belt 21 for conveying a plurality of containers 11, and these containers 11 have been equipped with the metal film 13 and the cover 14 shown in FIG. 1, that is, the metal film 13 After the port 12 of the container 11 is covered, and the container 11 is screwed on, the container 11 is placed on a conveyor belt 21 and transported in the direction of the arrow shown in FIG. 2.

The system 20 further includes a high-frequency generator 22, an infrared camera 23, a processing device 24, an output device (such as a display 25), and a removal device 26. First, the container 11 passes the high-frequency generator 22, and the high-frequency generator 22 applies the high-frequency to the container 11. Specifically, the high-frequency generator 22 applies the high-frequency to the position of the metal thin film 13 covered on the port 12 of each container 11. The applied high frequency causes an eddy current to be generated on the metal thin film 13 and generates heat to emit infrared radiation.

Shortly after the high frequency is applied to the container 11, the container 11 enters the field of view of the infrared camera 23. The infrared camera 23 located above the container 11 senses the infrared radiation emitted by the container 11 and generates the parts of the container 11 Temperature data. It can be seen from the configuration of the infrared camera 23 in FIG. 2 that the temperature data acquired by the infrared camera 23 is mainly contributed by the entire plane of the metal thin film 13.

The processing device 24 is coupled to the infrared camera 23, and the temperature data generated by the infrared camera 23 is transmitted to the processing device 24. The processing device 24 will set the temperature The data is converted into image data, and image processing is performed on the image data. The processed image can be displayed on the display 25. The processing device 24 also judges whether there is a heat-sealing defect based on the acquired image, for example, a crack that appears when the metal thin film 13 is not adhered well, and the detection result can also be displayed on the display 25.

When a heat seal defect is detected, the processing device 24 may issue a warning signal to notify the output device to issue a warning, such as a warning sound, a warning message on the display 25, etc. The processing device 24 may also control the removal device coupled thereto. 26. The container 11 with the heat-sealing defect is removed from the conveyor belt 21. For example, the removal device 26 has a movable arm which can be controlled by the processing device 24. When the arm is extended, the arm touches the container 11 and thereby pushes the container 11 away from the conveyor belt 21 and drops to a poor seal Collection area.

Please refer to FIG. 3, which illustrates a method for detecting a heat seal defect according to an embodiment of the present disclosure. In the heat seal detection performed by the processing device 24, the present disclosure improves the conventional detection method, and proposes a heat seal defect detection method that can improve the detection accuracy. Please refer to FIG. 3 with reference to FIG. 2. The method for detecting a heat seal defect of the present disclosure includes the following steps.

Step S31: A high frequency is applied to the position of the metal thin film 13 covered on the port 12 of the container 11. In this step, the high frequency generator 22 generates a high frequency and applies it to the container 11. The generated high frequency is focused on the metal thin film 13 covered on the port 12 of the container 11, so that the metal thin film 13 generates heat and emits infrared radiation.

Step S32: Use an infrared camera 23 to sense infrared radiation above the container 11 to generate temperature data of each part of the container 11. In this step, The container 11 moves from the position of the high-frequency generator 22 to the lower portion of the infrared camera 23 through the conveyor belt 21 and enters the field of view of the infrared camera 23. At this time, the infrared camera 23 senses the infrared radiation emitted by the container 11 to generate the container 11 For the temperature data of each part, since the metal thin film 13 is a heat source, this temperature data is mainly contributed by the metal thin film 13, and because of the configuration of the sensing angle, the infrared camera 23 can sense the temperature of the entire flat surface of the metal thin film 13.

Step S33: Convert the temperature data of each part of the container into an image of the container. The processing device 24 receives the temperature data generated by the infrared camera 23 and converts the temperature data into an image. In the process, the temperature data may also be filtered, and the obtained image may be further image processed to optimize the Image, or simplify the image, so that the processing speed is accelerated, so that in the subsequent defect detection, the effect of real-time detection can be achieved.

Step S34: Position the image of the container to define a region of interest (ROI). Generally speaking, most of the areas where heat sealing defects occur are fixed. Using this feature, these areas can be located on the image using inspection, and only by looking at the temperature distribution of these areas, accurate detection can be achieved. This can also effectively reduce the time taken for defect detection. In terms of accuracy, only detecting these areas can also avoid the temperature data being affected by other areas where defects are unlikely to occur, and reduce the accuracy. For example, the port 12 of the container 11 often has a heat-sealing defect because it is not properly sealed. At this time, the area around the port 12 can be detected.

Step S35: Calculate the continuous area occupied by the temperature not within a qualified temperature range in the region of interest of the image. Generally, if the seal is The temperature distribution of the metal thin film 13 will be uniform. If the seal is not sure and a breach occurs, the temperature will drop significantly due to heat dissipation at the breach, and the temperature distribution of the metal thin film 13 will be uneven. If there is no heat dissipation, the temperature of each part of the metal thin film 13 will be basically in a temperature interval, that is, a qualified temperature range. If a breach occurs, the flow of heat causes the temperature of the metal thin film 13 to be lower in some places, while the temperature in a small part will increase. In this step, by analyzing the region of interest of the image of the container 11 and calculating the continuous area occupied by the temperature that is not within the acceptable temperature range, it can be determined whether there is a sealing defect. For example, through several tests, the acceptable temperature range is, for example, 33 to 40 ° C. At this time, the continuous area occupied by temperatures below 33 ° C and above 40 ° C can be calculated. In another embodiment, the continuous area can also be calculated for each temperature interval. After eliminating possible noise, the effective continuous area is added up. For example, for low temperature, calculate its continuous area from 33 ° C to 32 ° C, calculate its continuous area from 32 ° C to 31 ° C, and so on to a set low temperature value; for high temperature, calculate it from 40 ° C to 41 ° C Continuous area. Calculate the continuous area from 32 ° C to 31 ° C, and so on to a set high temperature value. It is also possible to make a defect judgment only for a low temperature situation, and a defect judgment for a high temperature situation alone, or to add the continuous areas of high temperature and low temperature situations together to make a defect judgment.

Step S36: Determine whether the continuous area is greater than an area threshold. In this step, an area threshold can be obtained through the test, which can be used as a criterion for judging the defect of the heat seal. When the continuous area obtained in step S35 is greater than the area threshold, it is determined that there is a heat seal defect; and when the continuous area obtained in step S35 is less than or equal to the area threshold, it is determined that there is no heat seal defect.

Step S37: Indicate a sealing defect in response to a case where the continuous area is greater than the area threshold. When it is determined in step S36 that there is a sealing defect, the processing device 25 may output an instruction signal to the output device. The output device can be a display 25, which can display a message on the screen that a sealing defect has occurred. The output device can also be a loudspeaker, which can issue a warning sound when there is a sealing defect. The processing device 25 may also control the output device (such as the removal device 26) to remove the container 11 in which the sealing defect occurs from the conveyor belt 21.

Please refer to FIG. 4, which illustrates another method for detecting a heat seal defect according to an embodiment of the present disclosure. In the method for detecting a heat seal defect shown in FIG. 4, steps S41 to S45 are the same as or similar to steps S31 to S35 in FIG. 3. For details, refer to the related description of steps S31 to S35.

Step S46: Calculate a difference between the standard deviation of the continuous area and the standard deviation of the region of interest or the standard deviation of the region after the continuous area is removed from the region of interest. In this step, the difference between the continuous area and the standard deviation of the gradation of the region of interest can be calculated, and the standard of the gradation of the region of the continuous area and the region of interest after the continuous area is removed can also be calculated. The difference between the two.

Step S47: Determine whether to indicate the sealing defect according to the magnitude of the difference. In this step, it can be judged whether the difference is greater than a threshold, and if so, it is judged that there is a sealing defect. The larger the difference between the two standard deviations, the larger the temperature difference between the two blocks, indicating that there is indeed a sealing defect; the smaller the difference between the two standard deviations, the smaller the temperature difference between the two blocks, indicating that the sealing defect may be possible. does not exist.

In the embodiment shown in FIG. 4, in the standard deviation calculation method, the standard of comparison is the standard deviation of the continuous area and the standard deviation of the region of interest (or the standard deviation of the region of the region of interest after the continuous area is removed). ), Because it compares with its own image, compared to using a fixed threshold (such as the area threshold in the embodiment shown in Figure 3) for judgment, this method considers its own image change, Significantly improved accuracy.

The embodiment of FIG. 3 and the embodiment of FIG. 4 described above can be executed separately to determine whether there is a sealing defect. In another embodiment, it is also possible to first compare the continuous area with an area threshold, remove regions with smaller continuous areas, and then perform standard deviation comparison, which is also feasible.

In the method for detecting a heat seal defect of the present disclosure, defect detection is performed for a region of interest, and it is determined whether a continuous area occupied by a temperature not within a qualified temperature range is greater than an area threshold and / or a standard deviation of the continuous area and the interest The standard deviation of the region or the difference of the standard deviation of the region after the continuous area is removed from the region of interest, so as to determine whether there is a sealing defect. Because the present disclosure uses continuous area to track the temperature distribution of sealing defects, the temperature data of the container is more accurately analyzed, and it can be closer to the expression form of sealing defects. Therefore, compared with the conventional sealing defect detection method, it is possible to obtain more For accurate detection results, it will not reduce the detection speed.

As an example, FIGS. 5A and 5B show images without a sealing defect, and FIGS. 6A to 6C show images with a sealing defect. Figure 5A In the image of Fig. 5B, the seal of port 12 is indeed without cracks, and there are obvious cracks in Figs. 6A to 6C, which are classified as seal defects.

The step of converting the temperature data into an image (ie, the above steps S33 and S43) may include the following steps: the data in the temperature data whose temperature is higher than a first temperature is represented by a first pixel value; the temperature in the temperature data is Data below a second temperature are represented by a second pixel value; and data in the temperature data whose temperature is between the first temperature and the second temperature are represented by different pixel values. That is, the temperature data obtained by the infrared camera 23 can be filtered. The image can be implemented as, for example, RGB color scale or gray scale. Taking RGB color gradation as an example, data points with temperatures above 45 ° C in the temperature data can be represented in bright red, and data points with temperatures below 30 ° C in this temperature data can be represented in black, with temperatures between 30 ° C and The 45 ° C data points are represented by different RGB pixel values. The advantage of temperature data filtering is that temperatures outside this temperature range may be judged as defects. Representing the defects with the same pixel value can better highlight the shape of the defect, and it also simplifies the image and simplifies subsequent image analysis. .

Please refer to FIG. 7, which is a schematic diagram showing a region of interest marked in an image of a container according to an embodiment of the present disclosure. Assuming that the port 12 of the container 11 is circular and the cover 14 is also circular, the area around the port 12 can be used as the region of interest, so the steps of defining the region of interest (ie, the above steps S34 and S44) can include the following Steps: define a center point O in the image of the container 11; define a first circle with the center point O as the center of the circle and a first radius value R1 as the radius; with the center point O as the center of the circle, a The second radius value R2 is a radius, which defines a second circle center; and The area between the first circle and the second circle is defined as the area of interest, such as the slanted area in FIG. 6. Since the area around the port 12 is usually an area where a sealing defect occurs, this area can be detected as an area of interest. The first radius R1 can be defined by the inner diameter of the port 12 of the container 11 and the second radius R2 can be determined by the cover 14 The outer diameter is defined, and the center of the circle can be found by using common image algorithms, so that the region of interest can be defined.

The step of calculating the continuous area (that is, the above steps S35 and S45) may include the steps of: calculating a first temperature interval that is not within the qualified temperature range, and a first continuous area occupied in a region of interest of the image; calculation A second temperature interval outside the qualified temperature range, a second continuous area occupied in the region of interest of the image; and the sum of the first continuous area and the second continuous area as the continuous area. That is, as the above example, if the qualified temperature range is 33 to 40 ° C, 33 ° C to 32 ° C can be used as the first temperature interval, and 32 ° C to 31 ° C can be used as the second temperature interval. Their continuous area addition is used as a step. Continuous area in S35 and S45. In another embodiment, it may also be determined whether the continuous area of a certain temperature interval is greater than the area threshold to determine whether there is a sealing defect.

The step of calculating the continuous area (that is, the above steps S35 and S45) may include the steps of: dividing the temperature not within the qualified temperature range into a plurality of temperature intervals; calculating each temperature interval occupying a region of interest in the image The continuous area occupied by each temperature interval; determine whether the continuous area occupied by each temperature interval is greater than a threshold value; remove the temperature interval corresponding to the continuous area less than the threshold value; and the continuous area occupied by the temperature interval where the continuous area is greater than the threshold value Sum up. That is, as mentioned above For example, when the continuous area corresponding to a certain temperature interval is too small and may be noise, it can be eliminated. Similarly, it is also possible to simply judge the continuous area added after removing noise in certain temperature intervals, and determine whether it is greater than the area threshold, thereby determining whether there is a sealing defect.

Please refer to FIG. 8, which is a schematic diagram showing another region of interest marked in an image of a container according to an embodiment of the present disclosure. During the sealing process, the metal thin film 13 may be put against the heat-insulating paper 15 and cause different temperature distribution. That is, when the metal thin film 13 is placed on the heat-insulating paper 15, the metal thin film 13 is not insulated The heat generated at a relatively high temperature can also be detected in this disclosure.

First, in the process of defining the region of interest, the circle formed by the port 12 may be used as the region of interest, which may include the following steps: defining a center point O in the image of the container 11; and using the center point O Is a circle center, and a predetermined radius value R1 is a radius, and a circle is defined. Here, the predetermined radius R1 may be defined by the inner diameter of the port 12.

In this example, the sealing defect detection may include the following steps: calculating an average temperature of the temperature data corresponding to the pixels in the circle; determining whether the average temperature is greater than a threshold value; and indicating a sealing defect in response Cases where the average temperature is greater than this threshold. In this type of detection, the average temperature can be used for calculation. This calculation can be achieved by averaging the temperatures corresponding to all pixels in the region of interest, or it can be calculated from the perspective of the total area. Of course, the above-mentioned concept of calculating the continuous area for a temperature range that is not within the acceptable temperature range can also be calculated.

Referring to FIG. 9, the region of interest can be divided into a plurality of unit regions 90. If one of the unit regions has a sealing defect, it is determined that the container 11 has a sealing defect. As shown in FIG. 9, the region of interest is divided into eight unit regions 90 at a predetermined angle. Each unit region 90 is judged separately. If there is a sealing defect in one unit region 90, it can be determined that the container 11 has a sealing defect. For each unit area 90, the judgment method shown in FIG. 3 and / or FIG. 4 can be adopted.

With reference to the above method and system for detecting heat seal defects, the present disclosure also proposes a device for detecting heat seal defects. The specific implementation thereof can be referred to the above method and system for detecting heat seal defects without further description.

The present disclosure has been disclosed as above with a preferred embodiment, but it is not intended to limit the present disclosure. Those with ordinary knowledge in the technical field to which this disclosure belongs can make various changes and modifications without departing from the spirit and scope of this disclosure. Therefore, the scope of protection of this disclosure shall be determined by the scope of the appended patent application.

Claims (17)

A method for detecting a heat-sealing defect, which is used to detect the sealing characteristic of a port of a container, the port of the container is covered with a metal film, and a cover is stacked on the metal film, the method includes: A high frequency is applied to the position of the metal film covered on the port of the container; an infrared camera is used to sense infrared radiation above the container to generate temperature data of each part of the container; the temperature data of each part of the container is converted into the container Positioning an image of the container to define a region of interest (ROI); calculating a temperature that is not within a qualified temperature range and a continuous area occupied by the region of interest of the image ; Calculate a difference between the standard deviation of the continuous area and the standard deviation of the region of interest or the standard deviation of the region after the continuous area is removed from the region of interest; and determine whether to indicate the seal defect based on the size of the difference . The method for detecting a heat-sealing defect as described in item 1 of the patent application scope, wherein the metal thin film includes an aluminum film. The method for detecting a heat-sealing defect as described in item 1 of the scope of the patent application, wherein the step of converting temperature data of each part of the container into an image of the container includes: data of the temperature data having a temperature higher than a first temperature , Represented by a first pixel value; data in which the temperature data is lower than a second temperature is represented by a second pixel value; and the temperature in the temperature data is between the first temperature and the second temperature The data is represented by different pixel values. The method for detecting a heat-sealing defect as described in item 1 of the scope of the patent application, further comprises: dividing the region of interest into a plurality of unit regions, wherein a temperature not within the qualified temperature range is calculated, and the region of interest of the image is calculated. The continuous area occupied by step includes: calculating the temperature outside the qualified temperature range, the continuous area occupied by each unit area, and when one of the unit areas is determined to have a sealing defect, It is determined that the container has a sealing defect. The method for detecting a heat-sealing defect as described in item 1 of the scope of patent application, wherein the step of positioning the image of the container to define the region of interest includes: defining a center point in the image of the container; The center point is a circle center, a first radius value is a radius, and a first circle is defined; the center point is a circle center, and a second radius value is a radius, a second circle center is defined; and the first circle is defined; The area between the shape and the second circle is defined as the area of interest. The method for detecting a heat-sealing defect as described in item 1 of the scope of patent application, wherein the step of calculating a temperature outside the qualified temperature range and occupying a continuous area in the region of interest of the image includes: calculating the temperature not in the qualified temperature A first temperature interval within the range, the first continuous area occupied in the region of interest of the image; a second temperature interval not within the qualified temperature range, calculated in the region of interest of the image A second continuous area; and a sum of the first continuous area and the second continuous area as the continuous area. The method for detecting a heat seal defect according to item 1 of the scope of the patent application, wherein the step of calculating a continuous area occupied by the temperature outside the qualified temperature range in the region of interest of the image includes: not at the qualified temperature The temperature in the range is divided into a plurality of temperature intervals; the continuous area occupied by each temperature interval in the region of interest of the image is calculated; the continuous area occupied by each temperature interval is greater than a threshold; the continuous area removed is less than A temperature interval corresponding to the threshold; and summing up the continuous area occupied by the temperature interval with a continuous area greater than the threshold. The method for detecting a heat-sealing defect as described in item 1 of the patent application scope, wherein the step of positioning the image of the container to define the region of interest includes: defining a center point in the image of the container; and A circle is defined by using the center point as a circle center and a predetermined radius value as a radius; and the method further includes: calculating an average temperature of temperature data corresponding to the pixels in the circle; and determining whether the average temperature is Greater than a threshold; and indicating a seal defect in response to a situation where the average temperature is greater than the threshold. A device for detecting a heat seal defect, which is used to detect the sealing characteristics of a port of a container, the port of the container is covered with a metal film, and a cover is stacked on the metal film. The device includes: a high frequency A generator for applying high frequency to the position of the metal film covered on the port of the container; an infrared camera for sensing infrared radiation above the container after the high frequency is applied by the high frequency generator to generate the Temperature data of each part of the container; a processing device coupled to the infrared camera for receiving the temperature data generated by the infrared camera; converting the temperature data of each part of the container into an image of the container; The image of the container is positioned to define a region of interest (ROI); the temperature is not within a qualified temperature range, and the continuous area occupied by the image's region of interest; the continuous area is calculated The difference between the standard deviation of the region of interest or the standard deviation of the region of interest after the continuous area is removed ; And according to the size of the difference, indicating that the seal is determined whether the defect; and an output means, coupled to the processing means for indicating the sealing defect. The apparatus for detecting a heat seal defect as described in item 9 of the scope of the patent application, wherein the metal thin film includes an aluminum film. The apparatus for detecting a heat seal defect as described in item 9 of the scope of patent application, wherein in the process of converting the temperature data into the image, the processing device is further configured to: The data is represented by a first pixel value; the temperature of the temperature data is lower than a second temperature by a second pixel value; and the temperature in the temperature data is between the first temperature and the second Data between temperatures is expressed as distinct pixel values. The device for detecting a heat seal defect as described in item 9 of the scope of the patent application, wherein the processing device is further configured to divide the region of interest into a plurality of unit regions; and calculate a temperature not within the qualified temperature range at each The continuous area occupied by the unit areas. When one of the unit areas is determined to have a sealing defect, the processing device determines that the container has a sealing defect. The device for detecting a heat seal defect as described in item 9 of the scope of patent application, wherein in the process of defining the region of interest, the processing device is further used to define a center point in the image of the container; use the center point Is a circle center, a first radius value is a radius, a first circle is defined; the center point is a circle center, and a second radius value is a radius, a second circle center is defined; and the first circle and the The area between the second circles is defined as the area of interest. The device for detecting a heat seal defect as described in item 9 of the scope of patent application, wherein in the process of calculating the continuous area, the processing device is further configured to calculate a first temperature interval that is not within the qualified temperature range. A first continuous area occupied in the region of interest of the; calculating a second temperature interval not in the qualified temperature range, a second continuous area occupied in the region of interest of the image; and the first continuous area The sum with the second continuous area is taken as the continuous area. The device for detecting a heat seal defect as described in item 9 of the scope of patent application, wherein in the process of calculating the continuous area, the processing device is further configured to divide the temperature not within the qualified temperature range into a plurality of temperature intervals; calculation The continuous area occupied by each temperature interval in the region of interest of the image; determining whether the continuous area occupied by each temperature interval is greater than a threshold; removing the temperature interval where the continuous area is less than the threshold; and for continuous areas For temperature intervals greater than this threshold, the continuous area occupied by them is added up. The device for detecting a heat seal defect as described in item 9 of the scope of patent application, wherein in the process of defining the region of interest, the processing device is further used to define a center point in the image of the container; and the center A point is a circle center, a predetermined radius value is a radius, a circle is defined, and the processing device is configured to calculate an average temperature of temperature data corresponding to the pixels in the circle; and determine whether the average temperature is greater than a threshold value, and The output device is used to indicate a sealing defect in response to a situation where the average temperature is greater than the threshold. A system for detecting a heat seal defect, which is used to detect the sealing characteristics of a port of a container, the port of the container is covered with a metal film, and a cover is stacked on the metal film. The system includes: a high frequency The generator is arranged on a production line. The production line has a conveyor belt for conveying a plurality of containers equipped with the metal film and the cover. The high frequency generator is used to cover the ports corresponding to each container. High frequency is applied to the position of the metal thin film; an infrared camera is used to sense infrared radiation above each container after the high frequency is applied by the high frequency generator to generate temperature data of each part of each container; a processing device Is coupled to the infrared camera to receive the temperature data generated by the infrared camera; convert the temperature data of each part of the container into an image of the container; position the image of the container to define a Region of interest (ROI); calculates the temperature not within a qualified temperature range, the interest in the image The continuous area occupied in the domain; calculating a difference between the standard deviation of the continuous area and the standard deviation of the region of interest or the standard deviation of the region after the continuous area is removed from the region of interest; and according to the magnitude of the difference To determine whether to indicate the sealing defect; and a removing device coupled to the processing device, in response to the processing device determining that the container has the sealing defect, and removing the container from the conveyor belt.