TWI703327B - Method of generating abnormal message and detection system - Google Patents

Abstract

A method of generating an abnormal message includes the following steps. Moving a carrying bed to a testing machine to obtain a mask image group. Providing a test object to the carrying bed, and capturing a test image of the carrying bed and the test object. Setting a detection area according to the mask image group. Then, calculating detected pixels corresponding to the detection area in the test image. If determining that the detected pixels meet a threshold setting condition, adjusting the carrying bed or the test object.

Description

Method and detection system for generating abnormal information

The present disclosure relates to a method for generating abnormal information, especially for scanning a detection object on a carrier.

Computed Tomography (Computed Tomography) is a detection technology that uses multiple X-rays to penetrate objects and reconstruct 3D images in the body through a computer. In addition to being applied outside the human body, it can also be applied to smaller organisms (such as mice).

When using computer tomography to detect organisms, the biology department is placed on a loading bed and sent to the machine for scanning. Therefore, it is necessary to confirm the correct placement of the creatures in order to avoid the scanning position error, which may cause the problem that the detection result cannot be interpreted.

One aspect of the present disclosure is a method for generating abnormal information, which includes the following steps: moving the carrier to the inspection machine to obtain a mask image group. Place the test object on the carrier. Capture the detection images of the bed and the detection object. Set the detection area according to the mask image group. Calculate the detection image The detection pixels in the detection area. Determine whether the detection pixel meets at least one abnormal threshold setting condition, and if so, adjust the bed or the detection object.

Another aspect of the present disclosure is a detection system including a carrier, a detection machine, an image detection device, and a processor. The carrier bed is used to carry the test object. The inspection machine is used to scan and inspect the carrier bed. In the case that no inspection object is placed on the carriage, the image detection device is used to capture the mask image group of the carriage. In the case that the test object is carried on the carrier, the image detector is used to capture the detected images of the carrier and the test object. The processor is used to generate the detection area according to the mask image group, and to calculate the detection pixels corresponding to the detection area in the detection image. The processor is also used to determine whether the detection pixel meets the abnormal threshold setting condition, and if so, an abnormal message is generated to adjust the bed or the detection object.

Accordingly, by calculating the detection pixels corresponding to the detection area in the detection image, it will be possible to determine whether the setting state of the loading bed is abnormal, so as to adjust the loading bed in real time to ensure that the detection machine scans the detection correctly.

100: detection system

110: Testing machine

111: mobile stand

111A: X-ray transmitter

112: mobile stand

112A: X-ray receiver

113: Processor

120: bed

121: Bed Cover

130: Image detection device

140: test object

150: storage unit

200: detection system

210: Testing machine

220: bed

221: Bed Cover

230: image detection device

240: test object

M10: Mask image group

M11: Initial image

M12: Unmasked image

M13: Convex mask image

M14: Skew mask image

D10: Inspection image

51: Second appearance image

71: The third appearance image

91: The fourth appearance image

320: Image

S401~S408: steps

S1001~S1009: steps

S1101~S1106: steps

D1: width

D2: width

R1: Detection area

R2: Detection area

R3: Detection area

Rt: image capture area

Fig. 1 is a schematic diagram of a detection system according to some embodiments of the present disclosure.

FIG. 2 is a schematic diagram of a testing machine according to some embodiments of the present disclosure.

FIG. 3 is a schematic diagram of a detection system according to some embodiments of the present disclosure.

Figure 4 is a diagram showing the difference according to some embodiments of the present disclosure Flow chart of the common message method.

Figures 5A to 5C are schematic diagrams of the mask generation process in some embodiments of the disclosure.

Figure 6 is a schematic diagram of an abnormality detection process in some embodiments of the disclosure.

Figures 7A-7C are schematic diagrams of the mask generation process in some embodiments of the disclosure.

Figure 8 is a schematic diagram of an abnormality detection process in some embodiments of the disclosure.

Figures 9A to 9C are schematic diagrams of the mask generation process in some embodiments of the disclosure.

Figure 10 is a schematic diagram of an abnormality detection process in some embodiments of the disclosure.

FIG. 11 is a schematic diagram of a detection system according to some embodiments of the present disclosure.

FIG. 12A is a schematic diagram of a carrier bed and a detection object according to some embodiments of the present disclosure.

Figure 12B is a schematic diagram of an image captured by the image detection device in some embodiments of the disclosure.

FIG. 13 is a flowchart of a method for generating an abnormal message according to some embodiments of the present disclosure.

FIG. 14 is a flowchart of a method for generating an abnormal message according to some embodiments of the present disclosure.

Hereinafter, multiple implementations of this case will be disclosed in schematic form. For the sake of clarity, many practical details will be described in the following description. However, it should be understood that these practical details should not be used to limit the case. In other words, in some implementations of the present disclosure, these practical details are unnecessary. In addition, in order to simplify the drawings, some conventionally used structures and elements will be shown in a simple schematic manner in the drawings.

In this text, when a component is referred to as “connected” or “coupled”, it can be referred to as “electrically connected” or “electrically coupled”. "Connected" or "coupled" can also be used to mean that two or more components cooperate or interact with each other. In addition, although terms such as “first”, “second”, etc. are used herein to describe different elements, the terms are only used to distinguish elements or operations described in the same technical terms. Unless clearly indicated by the context, the terms do not specifically refer to or imply order or sequence, nor are they used to limit the present invention.

Please refer to Figures 1 and 2, which are schematic diagrams of the detection system 100 in this disclosure. The detection system 100 includes a detection machine 110, a carrier 120 and an image detection device 130. Two mobile stands 111 and 112 are provided in the inspection machine 110. The mobile stands 111 and 112 relatively rotate in the inspection machine 110, and are respectively provided with an X-ray transmitting device 111A and an X-ray receiving device 112A for scanning and detecting the inside of the inspection machine 110. In some embodiments, the inspection machine 110 performs a computed tomography scan through the X-ray transmitter 111A and the X-ray receiver 112A, but the disclosure is not limited to this.

The carrier bed 120 is used to carry the test object 140, and can be sent to the test machine 110 through the conveying device for scanning test. In the department In this embodiment, the image detection device 130 (such as a photographic lens) is installed in the inspection machine 110, so that after the carrier 120 and the inspection object 140 are moved into the inspection machine 110, the image detection device 130 is used The appearance images of the carrier bed 120 and the detection object 140 are captured. In some embodiments, different specifications (eg, large, medium, or small) of the bed 120 have different methods for generating abnormal messages. The details will be detailed in the subsequent paragraphs.

Please refer to FIG. 3. In some embodiments, the detection system 100 further includes a processor 113 and a storage unit 150. The processor 113 is electrically connected to the image detection device 130 and the storage unit 150, and is used to perform calculations based on the image captured by the image detection device 130. The storage unit 150 is used to store the image captured by the image detection device 130 and at least one abnormal threshold setting condition 151. The storage unit 150 may be a hard disk in the testing machine 110, or may be an external computer.

Please refer to FIG. 4, which illustrates the flowchart of some embodiments of the method for generating an abnormal message of the present disclosure. The method for generating an abnormal message includes the following steps S401 to S408. In step S401, the processor 113 controls the conveying device (such as a conveying table) on the inspection machine 110 to move the carrier 120 to a preset position in the inspection machine 110. At this time, the detection object 140 is not yet carried on the carrier bed 120, and the image detection device 130 captures the appearance of the carrier bed 120 to generate the mask image group M10, and stores the mask image group M10 in the storage unit 150. In some embodiments, the inspection system 100 executes the "mask generation process" to generate the mask image group M10. The masked image group M10 includes at least one masked image, for example, an image of the bed 120 in an abnormal state (for example, no cover). In other embodiments, the mask image group M10 includes Multiple masked images. The method of generating the mask image group M10 will be detailed in subsequent paragraphs.

In step S402, after completing the "mask generation process", the processor 113 executes the "abnormality detection process" to set one or more detection areas according to the mask image group M10. The detection area is a special position of the carrier bed 120 in an abnormal state. For example, if the bed cover 121 on the carrier bed 120 is not installed correctly, the bed cover 121 may protrude from the carrier bed 120. Therefore, the detection area may be If an image is detected in the area above the bed 120, it means that an abnormality has occurred. The details of the detection area will be detailed in subsequent paragraphs.

In step S403, the conveying device on the testing machine 110 is controlled to move the bed 120 to the outside of the testing machine 110, and the user (such as the inspector) places the test object 140 on the bed 120. In step S404, the carrier bed 120 is moved to the same predetermined position in the inspection machine 110 again, and the image detection device 130 is used to capture the appearance of the carrier bed 120 and the inspection object 140 thereon as the inspection Image D10.

In step S405, the processor 113 calculates the detection pixels corresponding to the detection area in the detection image D10. In step S406, the processor 113 determines whether the detected pixel meets the abnormal threshold setting condition 151. If the result of the determination is yes, it means that the state of the bed 120 corresponding to the detected image D10 is incorrect. In step S407, the detection system 100 will Generate an abnormal message (for example, display a message such as "Bed cover not installed") to adjust the bed 120 or the detection object 140 on it according to the abnormal message. If not, it means that the loading bed 120 and the inspection object 140 on it have passed the inspection and the setting method is correct. In step S408, the inspection machine 110 will start to perform scanning.

Accordingly, since in step 401, the user captures the mask image group M10 for the carrier bed 120 in a known state. Therefore, before the user scans the inspection object 140 through the inspection machine 110, the user can use the aforementioned steps S403~S406, first obtain the detection image D10, and then calculate the detection image D10 corresponding to the detection pixels in the detection area to determine whether the setting of the loading bed 120 or the detection object 140 is abnormal, and to check the loading bed 120, Or adjust the amount or position of the detection object 140. The detection system 100 adjusts the loading bed 120 or the detection object 140 according to the detection pixels. The detection machine 110 can automatically determine whether the loading bed and the bed cover are properly closed, or whether the detection object is placed correctly and its size. An important component inside the object collision detection system 100.

In some embodiments, the calculation actions and judgment actions of the aforementioned steps S405 and S406 are executed by the processor 113 in the inspection machine 110, but the present disclosure is not limited thereto. In some other embodiments, the inspection machine 110 can also be connected to a server or an external computer to perform calculations through the server or the external computer.

In some embodiments, the abnormal threshold setting condition 151 may include a plurality of abnormal threshold values corresponding to different abnormal states. For example, in the process of the detection system 100 calculating the detection pixels of the detection image D10 corresponding to the detection area, the processor 113 is used to calculate the number of pixels in the detection pixels. If the number of pixels exceeds the abnormal threshold, it means that the state of the bed 120 corresponding to the detection image D10 is incorrect. At this time, the detection system 100 will generate an abnormal message (such as a message such as "No bed cover installed"), and the user can According to the abnormal information, readjust the amount or position of the detection object 140 on the bed 120, readjust the position of the bed cover on the bed 120, or set a bed on the bed 120 cover. The method for the detection system 100 to determine the detection pixels and the abnormal threshold setting condition 151 will be described in detail in subsequent paragraphs.

Here, the step M10 of obtaining the mask image group will be explained first. In some embodiments, the mask image group M10 includes an initial image M11 that "the bed 120 is in the correct setting state", and the processor 113 is used to generate the mask image group M10 corresponding to the initial image M11 according to the initial image M11 One or more masked images in abnormal state. For example, the detection system 100 will use the image detection device 130 to first obtain the appearances of "no cover", "convex cover" and "bed cover skewed" in the "mask generation process", and then combine these The difference between the appearance image and the initial image M11 is performed to obtain images of different abnormal states and set as the mask image in the mask image group M10.

Specifically, in some embodiments, the mask image group M10 includes an uncovered mask image M12, a convex mask image M13, and a skewed mask image M14. The uncovered mask image M12 corresponds to the appearance when the bed 120 does not have the bed cover 121 (such as the missing pixel area); the convex mask image M13 corresponds to the appearance when the bed cover 121 protrudes from the bed 120 ( Such as: the area of protruding pixels); the skewed mask image M14 corresponds to the appearance when the bed cover 121 and the carrier 120 maintain a skewed angle.

Here, according to the large, medium and small carrier beds, the corresponding detection methods are explained respectively. The large and medium-sized carrier bed 120 includes a bed cover 121. When the large and medium-sized carrier bed 120 is tested, the purpose is to determine whether the bed cover 121 is set correctly. The method for generating multiple mask images (ie, uncovered mask image M12, convex mask image M13, and skewed mask image M14) of the mask image group M10 is described here as follows.

As shown in FIGS. 1 and 5A to 5C, in the "mask generation process", when the carrier 120 is outside the inspection machine 110, the image detection device 130 captures the internal image of the inspection machine 110. Since there is no detection object 140 in the detection machine 110 at this time, the detected internal image should be a completely black screen.

Then, the carrier bed 120 is moved into the inspection machine 110, so that the image detection device 130 captures the first appearance image of the carrier bed 120 in the correct setting state, that is, the state in which the carrier bed 120 and the bed cover 121 are properly closed ( For example: in the center of the black screen, the area where the bed rests will be displayed in white pixels) The processor 113 of the inspection machine 110 compares the difference between the first external image and the internal image, and regards the internal image as a background color and removes it from the first external image to reduce unnecessary noise interference. The processor 113 sets the image obtained after the foregoing processing as an initial image M11 in the mask image group M10.

After the initial image M11 is generated, the bed cover 121 on the bed 120 is removed, and the bed 120 is moved to the inspection machine 110, so that the image detection device 130 captures the second when the bed 120 has no bed cover 121 Appearance image 51. In some embodiments, the image detection device 130 removes the same pixel area in the second external image 51 as the internal image (ie, removes the background color). Then, the difference between the second appearance image 51 and the initial image M11 is compared, and the difference calculation is performed to generate an unmasked image M12.

Please refer to FIG. 6, after completing the "mask generation process", the inspection system 100 will execute the "abnormality detection process" on the carrier 120 and the inspection object 140 to confirm whether the setting state of the carrier 120 is correct. In the "abnormality detection process", the inspection machine 110 moves the bed 120 and the inspection object 140 to In the detection machine 110, the image detection device 130 is used to capture the detection image D11. Please refer to Fig. 6, in some embodiments, after the image detection device 130 captures the detection image D11, it will first remove the area of the same pixel as the internal image from the detection image D11 (ie, remove the background color) ).

Then, as shown in FIG. 6, when the "abnormality detection process" is executed, the processor 113 of the inspection machine 110 will set the detection area R1 according to the area of the uncovered mask image M12 in the mask image group M10. The processor 113 will calculate the number of pixels in the detection image D11 corresponding to the detection area R1. In addition, when judging whether the detected pixel corresponds to the abnormal threshold setting condition 151, the processor 113 will determine whether the number of pixels is less than an abnormal threshold in the abnormal threshold setting condition 151 (e.g., 60% of the detection area R1 The area is white pixels). If it is less than the abnormal threshold value, it means that the carrier bed 120 has an abnormal state of "no cover", and a bed cover 121 should be installed on the carrier bed 120 according to the abnormal information.

The detection of another abnormal state is explained here. Please refer to FIGS. 7A to 7C. After the initial image M11 is generated in the same manner as the previous embodiment, the position of the bed cover 121 on the bed 120 is adjusted so that the bed cover 121 protrudes from the bed 120. Then, the carrier bed 120 is moved to the inspection machine 110, so that the image detection device 130 captures the third appearance image 71 of the bed cover 121 protruding from the carrier bed 120 (in some embodiments, the image detection device 130 Similarly, the area of the third external image 71 with the same pixel as the internal image will be removed). The difference between the third appearance image 71 and the initial image M11 is compared, and the difference calculation is performed to generate the convex mask image M13.

Then, as shown in Figure 8, when the "anomaly detection flow During the process, the processor 113 of the inspection machine 110 will set the inspection area R2 according to the area of the convex mask image M13 in the mask image group M10. The processor 113 will calculate the number of pixels in the detection image D12 corresponding to the detection area R2. In addition, when determining whether the detected pixel corresponds to the abnormal threshold setting condition 151, the processor 113 will determine whether the number of pixels is greater than another abnormal threshold in the abnormal threshold setting condition 151 (such as 10 in the detection area R2). % Area has white pixels). If it is greater than the abnormal threshold value, it means that the carrier bed 120 has an abnormal state of "convex cover". The position of the carrier bed 120 on the bed cover 121 should be adjusted according to the abnormal information.

Here is another abnormal state that may occur in the "medium bed". Please refer to FIGS. 9A to 9C. After the initial image M11 is generated in the same manner as the previous embodiment, the position of the bed cover 121 on the bed 120 is adjusted to maintain a skew angle between the bed cover 121 and the bed 120. Then, the carrier bed 120 is moved to the inspection machine 110, so that the image detection device 130 captures the fourth appearance image 91 of the carrier bed 120 and the bed cover 121 at a skew angle (in some embodiments, the image detection device 130 will remove the area of the fourth external image 91 with the same pixel as the internal image). When the bed cover 121 is skewed, the pixel area and distribution area of the fourth appearance image 91 will be significantly different from the initial image M11. By comparing the difference between the fourth appearance image 91 and the initial image M11 and performing a difference calculation, the skew mask image M14 can be generated.

As shown in FIG. 10, when the "abnormal detection process" is executed, the processor 113 of the inspection machine 110 will set the detection area R3 according to the area of the skewed mask image M14 in the mask image group M10. The processor 113 will calculate the number of pixels in the detection image D13 corresponding to the detection area R3. this In addition, when judging whether the detected pixel corresponds to the abnormal threshold setting condition 151, the processor 113 will determine whether the number of pixels is greater than another abnormal threshold in the abnormal threshold setting condition 151 (such as 10 in the detection area R3). % Area has white pixels). If it is greater than the abnormal threshold value, it means that the carrier bed 120 has an abnormal state of "skew", and the position of the bed cover 121 on the carrier bed 120 should be adjusted according to the abnormal information.

Accordingly, after the initial image M11, the uncovered mask image M12, the convex mask image M13, and the skewed mask image M14 are generated, the processor 113 of the inspection machine 110 can execute the "abnormal detection process". The abnormal state of the bed 120 is determined by comparing the mask image group M10 with the detection image D10. In some embodiments, when the "anomaly detection process" is executed, the processor 113 of the inspection machine 110 first performs the detection image D10 and the uncovered mask image M12, the convex mask image M13 and the mask image group M10. The skewed mask images M14 are sequentially compared to confirm which abnormal state the carrier 120 is in.

Please refer to Figure 11, which is a schematic diagram of the application of the method of generating abnormal messages in this disclosure to the "small-sized bed". In this embodiment, the inspection system 200 includes an inspection machine 210, a carrier 220, and an image detection device 230. The small carrier 220 does not have a bed cover, but carries the inspection object through the groove 221. When detecting "small-sized bed", the method of generating abnormal information can also include "mask generation process" and "abnormal detection process". In some embodiments, during the "mask generation process", the processor of the inspection machine 200 can use the image detection device 230 to capture the image of the carriage 220 in the correct state as the image in the mask image group The initial image.

In some embodiments, the image detection device 230 is set in the detection The measuring machine 210 is used to continuously detect the image of the image capturing area Rt until the carrier 220 completely passes through the image capturing area Rt. After the processor of the inspection machine 210 receives the image, it accumulates the image of the image capturing area Rt to generate a complete image corresponding to the bed 220. Similar to the previous embodiment, in some embodiments, the image detection device 230 removes the same pixel area as the internal image from the captured image (ie, removes the background color) to generate a corresponding carrier 220 initial image.

Please refer to FIGS. 12A and 12B. In some embodiments, the width D2 of the detection object 240 on the carrier 220 is greater than the width D1 of the carrier 220, and the image detection device 230 captures the image in the image capturing area Rt , And the accumulated result will be as shown in image 320 in Figure 12B. The aforementioned "accumulation" means that the inspection system 100 will capture the image outline. Therefore, although the inspection object 240 in Figure 9A only occupies the middle part of the bed 220, the lower half of the detected image still has Width D2. In other embodiments of the present disclosure, the image detection device 230 can generate an initial image, an uncovered mask image, a convex mask image, a skewed mask image, and a detection image through this method.

Since the small bed 220 does not have a bed cover, the mask image group includes the initial image M11, but does not need to include abnormal state mask images such as "no cover", "convex cover" or "bed cover skewed". In the "abnormal detection process", the image detection device 230 can capture the images of the bed 220 and the detection object 240 on it through the same principle shown in Figures 12A and 12B to obtain the detection image (ie, normal state). The images of the lower carrier 220 and the detection object 240 are shown in Figure 12A). The processor of the inspection machine 210 based on the initial image M11 Set the detection area and calculate the detection pixels corresponding to the detection area in the detection image.

In some embodiments, since the image detection device 230 continuously captures images in the image capture area Rt, if the processor is accumulating the images, it finds that the captured detection images correspond to the detection If the detection pixels of the area exceed the abnormal threshold value (for example, more than 10% are white pixels), the detection machine 210 can interrupt the accumulation process and directly generate an abnormal message. That is, the processor can calculate the detection pixels corresponding to the detection area in the detection image while accumulating the images in the image capture area Rt. For example, if the detection area is the "width D1" of the carrier 220, once the processor determines that the area or area width of the detection pixel is greater than the width D1, an abnormal message will be generated.

The large, medium and small carrier beds are organized here, and the detection steps are explained. Referring to FIGS. 1 and 13, in step S1001, the setting mode of the carrier bed 120 and the bed cover 121 is adjusted, and the carrier bed 120 and the bed cover 121 are moved into the inspection machine 110. In step S1002, the image detection device 130 captures images of the bed 120 and the bed cover 121 to generate a mask image in the mask image group M10.

In step S1003, it is determined whether all the mask images in the mask image group M10 (such as uncovered mask image M12, convex mask image M13, and skewed mask image M14) have been generated, and each mask image will be Respectively correspond to an abnormal threshold setting condition. If all the mask images are not stored in the storage unit 150, return to step S1001. If all the mask images have been stored in the storage unit 150, it means that the "mask generation process" has been completed, and the "abnormality detection process" can be executed.

In step S1004, the carrier bed 120 is moved outside the inspection machine 110, the inspection object 140 is placed on the carrier bed 120, and the carrier bed 120 is moved to the inspection machine 110 again, and the image detection device 130 captures the load. The images of the bed 120 and the bed cover 121 are used to generate a detection image. In step S1005, the detection area is set based on the mask image. In some embodiments, the uncovered mask image M12, the convex mask image M13, and the skewed mask image M14 can respectively correspond to a detection area. In step S1006, it is determined whether the detection pixels corresponding to the detection area of the unmasked image M12 in the detection image meet the abnormal threshold setting condition (eg, less than 60%), and if so, step S1009 is executed to generate an abnormal message.

In step S1007, it is determined whether the detection pixels corresponding to the detection area of the convex mask image M13 in the detection image meet the abnormal threshold setting condition (eg, greater than 20%). If yes, execute step S1009 to generate an abnormal message.

In step S1008, it is further determined whether the detection pixels corresponding to the detection area of the skew mask image M14 in the detection image meet the abnormal threshold setting condition (eg, greater than 20%). If yes, execute step S1009 to generate an abnormal message. If the judgments in step S1006 to step S1008 are all negative, it means that the state of the carrier 120 is correct, and the inspection machine 210 will perform scanning.

Please refer to Fig. 14, when the small carrier 220 is detected, in step S1101, the carrier 220 is moved toward the detection machine 210 so that the carrier 220 passes through the detection area R1. In step S1102, the image detection device 230 captures and accumulates the images of the loading bed 220 to generate the initial image in the mask image group M10. In step S1103, move the carrier 120 to the inspection machine Outside 110, a test object 240 is placed on the bed 220, and the bed 220 is moved toward the direction of the inspection machine 210. In step S1104, the image detection device 230 captures and accumulates images of the loaded bed 220 to generate a detection image D10. In step S1105, the detection area is set based on the initial image in the mask image group M10.

In step S1106, it is calculated whether the detection pixels corresponding to the detection area in the detection image D10 meet the abnormal threshold setting condition (eg, greater than 10%). If so, in step S1106, the inspection machine 210 will generate an abnormal message. On the contrary, if the state of the carrier bed 220 is normal, the inspection machine 210 will perform scanning.

The various elements, method steps, or technical features in the foregoing embodiments can be combined with each other, and are not limited to the order of description or presentation of figures in the present disclosure.

Although the present disclosure has been disclosed in the above embodiments, it is not intended to limit the content of the present invention. Anyone who is familiar with the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the present invention The scope of protection of the content shall be subject to the scope of the attached patent application.

S401~S408: steps

Claims (14)

A method for generating abnormal information, the steps include: moving a carrier bed to a detection machine to obtain a mask image group; moving the carrier bed outside the detection machine, and placing a test object on the carrier, Move the bed and the inspection object to the inspection machine again; capture an inspection image of the bed and the inspection object; set a inspection area according to the mask image group; calculate that the inspection image corresponds to the inspection A detection pixel of the area; and determine whether the detection pixel meets at least one abnormal threshold setting condition, and if so, adjust the carrier or the detection object. The method for generating an abnormal message according to claim 1, wherein the step of determining whether the detected pixel meets the abnormal threshold setting condition includes: determining whether the detected pixel is below an abnormal threshold; Under the condition that the abnormal threshold is lowered, an abnormal message is generated; and according to the abnormal message, a bed cover is set on the bed. The method for generating an abnormal message according to claim 1, wherein determining whether the detected pixel meets the abnormal threshold setting condition includes: determining whether the detected pixel exceeds an abnormal threshold; when the detected pixel exceeds the abnormal threshold In the case of an abnormal threshold, an abnormal message is generated; and According to the abnormal information, adjust the position of the test object or one of the bed covers. The method for generating an abnormal message according to claim 1, wherein the step of obtaining the mask image group includes: capturing an internal image of the inspection machine; capturing the correct fit between the carrier bed and a bed cover A first external image; and comparing the difference between the first external image and the internal image to generate an initial image. The method for generating an abnormal message as described in claim 4, wherein the step of obtaining the mask image group further includes: removing the bed cover of the bed; capturing a second when the bed is not covered An appearance image; and comparing the difference between the second appearance image and the initial image to generate an uncovered mask image in the mask image group. The method for generating an abnormal message according to claim 4, wherein the step of obtaining the mask image group further includes: adjusting the position of the bed cover on the bed so that the bed cover protrudes from the bed; capturing The bed cover protrudes from a third appearance image of the supporting bed; and the difference between the third appearance image and the initial image is compared to generate a convex mask image in the mask image group. The method for generating an abnormal message according to claim 4, wherein the step of obtaining the mask image group further comprises: adjusting the position of the bed cover on the bed so that the bed cover and the bed are kept at a skew angle ; Capture a fourth appearance image maintaining the skew angle between the bed cover and the carrier; and compare the difference between the fourth appearance image and the initial image to generate a skew mask in the mask image group image. The method for generating an abnormal message as described in claim 1, wherein the step of obtaining the mask image group includes: capturing an internal image of the inspection machine; making the bed pass an image capture in the inspection machine Capture area; capture and accumulate images in the image capture area to generate an external image; compare the difference between the external image and the internal image as an initial image. The method for generating an abnormal message as described in claim 8, further comprising: passing the carrier bed and the inspection object through the image capturing area in the inspection machine; and capturing and accumulating the images in the image capturing area, To produce the detection shadow Like. The method for generating an abnormal message according to claim 1, wherein the mask image group includes a plurality of mask images, and the at least one abnormal threshold setting condition includes a plurality of abnormal threshold setting conditions corresponding to the mask images , To determine whether the detection pixel meets the abnormal threshold setting conditions, if none of them, the detection machine executes scanning. A detection system includes: a loading bed for carrying a detection object; a detection machine for scanning and testing the loading bed; and an image detection device, when the detection object is not placed on the loading bed Next, the image detection device is used to capture a set of masked images of the bed; when the inspection object is carried on the bed, the image detector is used to capture the image of the bed and the inspection object A detection image; and a processor for generating a detection area according to the mask image group, and for calculating a detection pixel of the detection image corresponding to the detection area, and the processor is also used for determining the detection pixel Whether it meets an abnormal threshold setting condition, if so, an abnormal message is generated to adjust the bed or the detection object. The detection system according to claim 11, wherein the processor is used to calculate the number of pixels in the detection pixel to determine whether the number of pixels exceeds an abnormal threshold. The detection system according to claim 11, further comprising: a storage unit electrically connected to the processor and used for storing the mask image group; wherein when the carrier bed is in a correct state, the image The detection device is used to capture an internal image of the inspection machine, and capture a first appearance image when the carrier bed and a bed cover are properly closed, so that the processor is used to compare the first appearance image with The difference between the internal images is used to generate an initial image. The detection system according to claim 13, wherein the mask image group includes an uncovered mask image, a convex mask image, and a skewed mask image; the uncovered mask image corresponds to that the bed does not have the bed The appearance of the cover; the convex cover mask image corresponds to the appearance of the bed cover protruding from the bed; the skewed mask image corresponds to the appearance of the bed cover and the bed at a skew angle.

Images