Inspection system of textile machine

Abstract

To provide an inspection system of a textile machine capable of acquiring a distribution of a degree of abnormality of the textile machine while reducing the data capacity stored in a storage section.SOLUTION: An inspection system of a textile machine includes: a classification section that detects a degree of abnormality of the textile machine on the basis of a newly acquired captured image and classifies the newly acquired captured image into one of a plurality of groups on the basis of the detected degree of abnormality of the textile machine; a storage section that stores the captured image for each group; a counting section that counts the number of captured images classified into a target group, which is a group into which the newly acquired captured image is classified, and stores the counted number of captured images in the storage section; and a selection section that determines whether or not to replace the captured image of the target group stored in the storage section with the newly acquired captured image when the number of captured images of the target group stored in the storage section reaches a predetermined upper limit number.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present invention relates to an inspection system for a textile machine. [Background technology]

[0002] Patent Document 1 discloses a loom shedding defect detection device that detects warp shedding defects. The loom shedding defect detection device includes a camera that photographs the warp shedding and a control device that detects shedding defects based on the image captured by the camera. The photographed image is stored in a memory unit of the control device.

[0003] The inspection system described in Patent Document 2 includes an imaging unit that images a workpiece, a storage unit that stores the image captured by the imaging unit, a first judgment unit that determines whether the workpiece shown in the image is defective, and a second judgment unit that determines whether the image should be stored in the storage unit. The distribution of measurement values ​​in the inspection of a workpiece previously determined to be a non-defective by the first judgment unit is divided into N regions. If the measurement value measured in the inspection belongs to a region of the N regions where the number of saved images is less than a specified number, the second judgment unit stores the image in the storage unit. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2020-196972 [Patent Document 2] Patent Publication No. 2021-144000 Summary of the Invention [Problem to be solved by the invention]

[0005] In Patent Document 1, when a camera continuously captures images while a loom is in operation, the number of captured images becomes enormous. Therefore, if all captured images are to be saved, the storage capacity of the storage unit is strained. For example, as in Patent Document 2, it is conceivable to reduce the amount of data stored in the storage unit by setting an upper limit on the number of captured images to be saved. However, in this case, the number of captured images in each region is unknown, and therefore the distribution of captured images across the entire region cannot be obtained. Therefore, it becomes difficult to grasp the state of the inspection target. [Means for solving the problem]

[0006] a sorting unit that determines whether the number of images of the target group stored in the storage unit has reached a predetermined upper limit, and if it determines that the number of images of the target group stored in the storage unit has not reached the predetermined upper limit, stores the newly acquired image in the storage unit as the image of the target group; and if it determines that the number of images of the target group stored in the storage unit has reached the predetermined upper limit, determines whether to replace the image of the target group stored in the storage unit with the newly acquired image.

[0007] According to the above configuration, when the number of photographed images of the target group stored in the memory unit has reached a predetermined upper limit, the sorting unit determines whether to replace the photographed images of the target group stored in the memory unit with newly acquired photographed images. This limits the number of photographed images of each group stored in the memory unit so as not to exceed the upper limit, thereby reducing the amount of data stored in the memory unit. Furthermore, the counting unit counts the number of photographed images classified into the target group and stores the counted number of photographed images in the memory unit. This makes it possible to obtain the distribution of abnormality levels of the textile machine without having to save all photographed images. Therefore, it is possible to obtain the distribution of abnormality levels of the textile machine while reducing the amount of data stored in the memory unit.

[0008] In the above-mentioned textile machine inspection system, the classification unit may calculate the degree of abnormality of the textile machine as a score, and classify the captured image into one of the multiple groups based on the calculated score.

[0009] According to the above configuration, the degree of abnormality of the textile machine can be detected more precisely than when the degree of abnormality of the textile machine is detected in stages such as S, A, B, C or excellent, good, fair, unacceptable.

[0010] In the above-mentioned textile machine inspection system, the classification unit calculates the score so that the higher the degree of abnormality of the textile machine, the lower the score, and if the score of the newly acquired captured image is lower than the score of the captured image with the highest score among the captured images of the target group stored in the memory unit, the selection unit may replace the captured image with the highest score among the captured images of the target group stored in the memory unit with the newly acquired captured image.

[0011] According to the above configuration, captured images with low scores, i.e., captured images with a high degree of abnormality in the textile machine, are preferentially stored in the storage unit, making it easier to use the captured images when investigating the cause of the abnormality in the textile machine, compared to when captured images with high scores, i.e., captured images with a low degree of abnormality in the textile machine, are preferentially stored in the storage unit.

[0012] In the above-described textile machine inspection system, the classification unit may further classify the captured image into one of the plurality of groups based on the capture time of the captured image. According to the above configuration, it is possible to acquire the time transition of the distribution of the abnormality degree of the textile machine.

[0013] The textile machine inspection system may further include a display unit, which displays a table indicating the number of captured images classified into each group. According to the above configuration, the operator can easily grasp the distribution of abnormality levels in the textile machine.

[0014] In the above-described textile machine inspection system, the display unit may display the table in a heat map format according to the number of the captured images classified into each group. According to the above configuration, the operator can more easily grasp the distribution of abnormality levels in the textile machine. [Effects of the Invention]

[0015] According to the present invention, it is possible to acquire the distribution of abnormality degrees of a textile machine while reducing the amount of data stored in the storage unit. [Brief explanation of the drawings]

[0016] [Figure 1] FIG. 1 is a block diagram showing the configuration of an inspection system for a textile machine. [Figure 2] FIG. 2 is a schematic diagram showing an example of an image captured by a camera. [Figure 3] FIG. 3 is a block diagram showing the functional elements of the control device. [Figure 4]FIG. 4 is a flowchart showing the processing performed by the control device. [Figure 5] FIG. 5 is a diagram illustrating an example of data stored in the auxiliary storage device. [Figure 6] FIG. 6 is a schematic diagram showing an example of a table displayed on the display unit. [Figure 7] FIG. 7 is a schematic diagram showing an example of a captured image displayed on the display unit. DETAILED DESCRIPTION OF THE INVENTION

[0017] An embodiment of an inspection system for a textile machine will be described below with reference to Figures 1 to 7. The inspection system for a textile machine of this embodiment is applied to a loom as a textile machine. In the following description, the inspection system for a textile machine will be simply referred to as the "inspection system."

[0018] <Loom> As shown in FIG. 1, the loom 100 includes a loom base 101. The loom base 101 has a weft insertion nozzle and a main shaft 102 (not shown). The weft insertion nozzle is provided at a first end of the loom base 101 in the width direction. The loom 100 of this embodiment is an air jet loom. The weft insertion nozzle inserts a weft into an shed Ta (see FIG. 2) of warp threads T by ejecting air from the first end to the second end of the loom base 101 in the width direction. The main shaft 102 is driven to rotate by a motor (not shown). The timing at which the weft insertion nozzle inserts the weft corresponds to the rotation angle of the main shaft 102. Specifically, the weft insertion nozzle inserts the weft when the main shaft 102 has rotated a predetermined first angle from a predetermined reference angle.

[0019] The loom 100 includes an angle detection unit 103 and a signal output unit 104. The angle detection unit 103 detects the rotation angle of the main shaft 102. The angle detection unit 103 is, for example, a rotary encoder. The angle detection unit 103 is connected to the signal output unit 104. The angle detection unit 103 transmits the detected rotation angle of the main shaft 102 to the signal output unit 104. The signal output unit 104 outputs a signal to the control unit 13 of the inspection system 10 every time the rotation angle of the main shaft 102 received from the angle detection unit 103 reaches a reference angle.

[0020] <Inspection system configuration> The inspection system 10 includes an imaging device 10a and a computer 10b. The imaging device 10a includes a camera 11, lighting 12, and a control unit 13. The computer 10b includes a control device 14, an auxiliary storage device 15 as a storage unit, an input unit 16, and a display unit 17. The inspection system 10 inspects the loom 100 based on images captured by the camera 11.

[0021] For example, if some of the warp threads T protrude into the shed Ta (see FIG. 7), an abnormality has occurred in the loom 100. For example, the greater the amount of warp threads T protruding into the shed Ta, or the greater the number of warp threads T protruding into the shed Ta, the greater the degree of abnormality in the loom 100. The inspection system 10 of this embodiment inspects the loom 100 based on the warp threads T that appear in the captured image.

[0022] The camera 11 is a digital camera. The camera 11 has an imaging element. Examples of the imaging element include a CCD image sensor (Charge Coupled Device image sensor) and a CMOS image sensor (Complementary Metal Oxide Semiconductor image sensor).

[0023] The camera 11 is provided at a position where it can photograph the shed Ta of the warp thread T. Specifically, the camera 11 is disposed at a first end, which is the end on the side where the weft insertion nozzle is located in the width direction of the machine frame 101. The camera 11 photographs the shed Ta of the warp thread T from the upstream side to the downstream side in the weft insertion direction. The image photographed by the camera 11 has information on the time of photographing added thereto by a timestamp function.

[0024] Fig. 2 is a schematic diagram showing an example of an image P captured by the camera 11. The captured image P shown in Fig. 2 captures an shed Ta for warp threads T. In the captured image P shown in Fig. 2, the warp threads T do not protrude into the shed Ta.

[0025] The lighting 12 in this embodiment is an LED light. The lighting 12 is provided at a position where it illuminates the opening Ta of the warp thread T. Specifically, the lighting 12 is disposed at a first end, which is the end on the side where the weft insertion nozzle is located in the width direction of the loom 101. The lighting 12 illuminates the opening Ta of the warp thread T from the upstream side to the downstream side in the weft insertion direction.

[0026] As shown in FIG. 1, the control unit 13 includes a processor 13a and a main memory 13b. The processor 13a may be, for example, a central processing unit (CPU), a graphics processing unit (GPU), or a digital signal processor (DSP). The main memory 13b includes a random access memory (RAM) and a read-only memory (ROM). The main memory 13b stores program code or instructions configured to cause the processor 13a to execute processing. The main memory 13b, i.e., the computer-readable medium, includes any available medium accessible by a general-purpose or special-purpose computer. The control unit 13 may be configured with a hardware circuit such as an application-specific integrated circuit (ASIC) or a field-programmable gate array (FPGA). The control unit 13, which is a processing circuit, may include one or more processors operating according to a computer program, one or more hardware circuits such as ASICs or FPGAs, or a combination thereof.

[0027] The control unit 13 is connected to the camera 11 and the lighting 12. The control unit 13 controls the timing at which the camera 11 takes a photograph and the timing at which the lighting 12 emits light. The control unit 13 is connected to the signal output unit 104 of the loom 100. The signal output by the signal output unit 104 is input to the control unit 13 as an external input signal. By receiving the external input signal, the control unit 13 recognizes the timing at which the rotation angle of the main shaft 102 becomes a reference angle. The control unit 13 outputs a trigger signal to the camera 11 and the lighting 12 simultaneously every time the main shaft 102 rotates by a predetermined second angle from the reference angle.

[0028] The camera 11 takes an image when a trigger signal is input from the control unit 13. The lighting 12 emits light when a trigger signal is input from the control unit 13. Since the trigger signals are input to the camera 11 and the lighting 12 simultaneously, the lighting 12 illuminates the shed Ta of the warp threads T at the same time that the camera 11 takes an image of the shed Ta of the warp threads T.

[0029] The control device 14 includes a processor 14a and a main memory 14b. The processor 14a may be, for example, a CPU, a GPU, or a DSP. The main memory 14b includes RAM and ROM. The main memory 14b stores program code or instructions configured to cause the processor 14a to execute processing. The main memory 14b, i.e., the computer-readable medium, includes any available medium accessible by a general-purpose or special-purpose computer. The control device 14 may be configured with a hardware circuit such as an ASIC or FPGA. The control device 14, which is a processing circuit, may include one or more processors operating according to a computer program, one or more hardware circuits such as an ASIC or FPGA, or a combination thereof.

[0030] The control device 14 is connected to an auxiliary storage device 15, an input unit 16, and a display unit 17. The auxiliary storage device 15 is a non-volatile storage device that is rewritable. The auxiliary storage device 15 is, for example, a hard disk drive or a solid state drive. The input unit 16 is, for example, a keyboard or a mouse. The display unit 17 is a display. The control device 14 is also connected to the camera 11 of the photographing device 10a. The control device 14 acquires images captured by the camera 11. In this embodiment, the control device 14 acquires an image from the camera 11 each time the camera 11 captures an image. In other words, when the camera 11 captures an image once, the control device 14 acquires one captured image.

[0031] 3, the control device 14 has a classification unit 41, a counting unit 42, and a selection unit 43. The classification unit 41, the counting unit 42, and the selection unit 43 are functional elements that function when the processor 14a executes a program stored in the main memory unit 14b.

[0032] Each time a new image is acquired from the camera 11, the classification unit 41 detects the degree of abnormality of the loom 100 based on the newly acquired image. In this embodiment, the classification unit 41 assigns the detection result of the degree of abnormality of the loom 100 to the image. The classification unit 41 classifies the newly acquired image into one of a plurality of groups based on the detected degree of abnormality of the loom 100.

[0033] The auxiliary storage device 15 stores the captured images for each group. The counting unit 42 is configured to be able to count the number of classified photographed images for each group. The counting unit 42 counts the number of photographed images classified into the target group, which is the group into which the newly acquired photographed image has been classified. The counting unit 42 stores the counted number of photographed images, i.e., the number of photographed images classified into the target group, in the auxiliary storage device 15.

[0034] The selection unit 43 is configured to be able to determine whether the number of captured images of each group stored in the auxiliary storage device 15 has reached a predetermined upper limit number set for each group. The selection unit 43 determines whether the number of captured images of the target group stored in the auxiliary storage device 15 has reached a predetermined upper limit number. In this embodiment, the predetermined upper limit number is set to five for each group. Therefore, the selection unit 43 of this embodiment determines whether the number of captured images of the target group stored in the auxiliary storage device 15 has reached five.

[0035] If the selection unit 43 determines that the number of photographed images of the target group stored in the auxiliary storage device 15 has not reached a predetermined upper limit, it stores a newly acquired photographed image as a photographed image of the target group in the auxiliary storage device 15. If the selection unit 43 determines that the number of photographed images of the target group stored in the auxiliary storage device 15 has reached a predetermined upper limit, it determines whether or not to replace the photographed image of the target group stored in the auxiliary storage device 15 with the newly acquired photographed image. If the predetermined upper limit is two or more, the selection unit 43 determines whether or not to replace one of the photographed images of the target group stored in the auxiliary storage device 15 with the newly acquired photographed image.

[0036] <Processing performed by the control device 14> The following describes the processing performed by the control device 14. In this embodiment, the control device 14 performs the following processing each time a new photographed image is acquired from the camera 11.

[0037] As shown in FIG. 4, in step S1, the classification unit 41 detects the degree of abnormality of the loom 100 based on the warp threads T captured in a newly captured image acquired by the camera 11. In this embodiment, the classification unit 41 calculates the degree of abnormality of the loom 100 as a score. The score is calculated, for example, in the range of 0 to 1 to four decimal places. The classification unit 41 has AI (artificial intelligence) that has learned from images captured when the loom 100 is in a normal state as reference images. The classification unit 41 calculates a score closer to 1 the more similar the image captured by the camera 11 is to the reference image. Therefore, the classification unit 41 calculates the score so that the higher the degree of abnormality of the loom 100, the lower the score. For example, if the captured image is the captured image P shown in FIG. 2, the classification unit 41 calculates the score of the loom 100 as 0.9998 based on the similarity between the reference image and the captured image P.

[0038] In step S2, the classification unit 41 classifies the newly acquired captured image from the camera 11 into one of a plurality of groups based on the degree of abnormality of the loom 100 detected in step S1. The classification unit 41 of the present embodiment classifies the newly acquired captured image from the camera 11 into one of a plurality of groups based on the score of the loom 100 calculated in step S1 and the time at which the captured image was taken.

[0039] As shown in Figures 5 and 6, in this embodiment, the score is divided into multiple score ranges in increments of 0.01, thereby setting multiple groups based on the score. In the example shown in Figures 5 and 6, the classification unit 41 classifies the captured images into groups based on the score, based on the third decimal place of the score of the loom 100. For example, if the score of the loom 100 is 0.999 to 0.995, the classification unit 41 classifies the captured images into the 1.00-0.99 group. For example, if the score of the loom 100 is 0.994 to 0.990, the classification unit 41 classifies the captured images into the 0.99-0.98 group.

[0040] Furthermore, the shooting time of the camera 11 is divided into multiple shooting times in one-minute increments, thereby setting multiple groups based on the shooting times. The multiple groups into which the captured images are classified are made up of a combination of multiple groups based on the scores and multiple groups based on the shooting times. When M groups based on the scores are set and N groups based on the shooting times are set, M×N groups into which the captured images are classified are set. Note that M and N are each integers of 2 or greater.

[0041] The classification unit 41 classifies the captured image into a group among the multiple groups that corresponds to both the score of the loom 100 and the capture time of the captured image. For example, in the case of an image with a score of 0.9998 and captured at 8:01, the classification unit 41 classifies the captured image into group G11, which has a score range of 1.00-0.99 and a capture time of 8:01.

[0042] 4, in step S3, the counting unit 42 counts the number of captured images classified into the target group. Specifically, the counting unit 42 increments the number of captured images classified into the target group by one from the number counted previously. Also, in step S3, the counting unit 42 stores the counted number of captured images, i.e., the number of captured images classified into the target group, in the auxiliary storage device 15.

[0043] In step S4, the selection unit 43 determines whether or not the number of captured images stored in the auxiliary storage device 15 for the target group has reached a predetermined upper limit number. If the selecting unit 43 determines that the number of captured images stored in the auxiliary storage device 15 for the target group has not reached a predetermined upper limit (NO in step S4), the process proceeds to step S5. In step S5, the selecting unit 43 stores the newly acquired captured images in the auxiliary storage device 15 as captured images for the target group. Therefore, if the number of captured images for the target group stored in the auxiliary storage device 15 is four or less, the selecting unit 43 stores the newly acquired captured images in the auxiliary storage device 15 as captured images for the target group.

[0044] If the selecting unit 43 determines that the number of captured images stored in the auxiliary storage device 15 for the target group has reached a predetermined upper limit (YES in step S4), the process proceeds to step S6. In step S6, the selecting unit 43 determines whether or not to replace the captured images stored in the auxiliary storage device 15 as the captured images of the target group with newly acquired captured images. Therefore, if the number of captured images of the target group stored in the auxiliary storage device 15 is five, the selecting unit 43 determines whether or not to replace one of the captured images of the target group stored in the auxiliary storage device 15 with the newly acquired captured image.

[0045] If the score of the newly acquired photographed image is lower than the score of the photographed image with the highest score among the photographed images of the target group stored in the auxiliary storage device 15 (YES in step S6), the selection unit 43 of this embodiment proceeds to step S7. In step S7, the selection unit 43 replaces the photographed image with the highest score among the photographed images of the target group stored in the auxiliary storage device 15 with the newly acquired photographed image. That is, the selection unit 43 deletes the photographed image with the highest score among the photographed images of the target group stored in the auxiliary storage device 15 from the auxiliary storage device 15, and stores the newly acquired photographed image in the auxiliary storage device 15 as a new photographed image of the target group.

[0046] In this embodiment, the selection unit 43 ends the flow if the score of the newly acquired photographed image is equal to or higher than the score of the photographed image with the highest score among the photographed images of the target group stored in the auxiliary storage device 15 (NO in step S6). That is, the selection unit 43 does not replace the photographed image with the highest score among the photographed images of the target group stored in the auxiliary storage device 15 with the newly acquired photographed image. The selection unit 43 deletes the newly acquired photographed image without storing it in the auxiliary storage device 15.

[0047] As shown in FIG. 5, the auxiliary storage device 15 stores the photographed images in each group and the number of photographed images classified into each group. The number of marks at the top of each group in Fig. 5 indicates the number of captured images stored in the auxiliary storage device 15. As described above, when the number of captured images of the target group stored in the auxiliary storage device 15 reaches the upper limit, the selection unit 43 determines whether to replace the captured images of the target group stored in the auxiliary storage device 15 with newly acquired captured images. In other words, when the number of captured images of the target group stored in the auxiliary storage device 15 reaches the upper limit, the selection unit 43 does not store newly acquired captured images in the auxiliary storage device 15 by adding them to the captured images of the target group stored in the auxiliary storage device 15. Therefore, the maximum number of captured images of each group stored in the auxiliary storage device 15 in this embodiment is five.

[0048] 5 indicates the number of captured images classified into that group. Since counting unit 42 counts the number of classified captured images regardless of the number of captured images stored in auxiliary storage device 15, the number of captured images classified into each group can be five or more.

[0049] 6, the control device 14 creates a table H indicating the number of captured images classified into each group. The control device 14 of this embodiment creates the table H in the form of a heat map according to the number of captured images classified into each group.

[0050] For example, in Table H, groups with 120 or more but less than 150 classified images are displayed in light green. Groups with 150 or more but less than 175 classified images are displayed in dark green. Groups with 175 or more but less than 200 classified images are displayed in light red. Groups with 200 or more but less than 250 classified images are displayed in red. Groups with 250 or more classified images are displayed in dark red. Note that Table H shown in Figure 6 expresses the format of a heat map by using denser dot hatching for groups with a larger number of classified images.

[0051] The control device 14 displays the created table H on the display unit 17. The display unit 17 displays the table H indicating the number of captured images classified into each group. In this embodiment, the display unit 17 displays the table H in a heat map format according to the number of captured images classified into each group.

[0052] The worker uses the input unit 16 to select a group for which the worker wishes to check the photographed images from table H displayed on the display unit 17. The control device 14 causes the display unit 17 to display the photographed images of the group selected by the worker via the input unit 16 from among the photographed images of each group stored in the auxiliary storage device 15. The control device 14 of this embodiment causes the display unit 17 to display the score of the loom 100 together with the photographed images.

[0053] Fig. 7 is a schematic diagram showing an example of a captured image P displayed on the display unit 17. In Fig. 7, five captured images P are displayed on the display unit 17. In Fig. 7, the score of the loom 100 is displayed below the captured image P.

[0054] [Operation of this embodiment] The operation of this embodiment will be described. The inspection system 10 that inspects the loom 100 based on images captured by the camera 11 includes a classification unit 41, an auxiliary storage device 15, a counting unit 42, and a sorting unit 43.

[0055] The classification unit 41 detects the degree of abnormality of the loom 100 based on the newly acquired photographed image every time a photographed image is newly acquired from the camera 11. The classification unit 41 classifies the newly acquired photographed image into one of a plurality of groups based on the detected degree of abnormality of the loom 100. The auxiliary storage device 15 stores the photographed images by group.

[0056] The counting unit 42 counts the number of photographed images classified into the target group, which is the group into which the newly acquired photographed image is classified. The counting unit 42 stores the counted number of photographed images, i.e., the number of photographed images classified into the target group, in the auxiliary storage device 15. This makes it possible to obtain the distribution of the abnormality degree of the loom 100 without saving all photographed images.

[0057] The selection unit 43 determines whether the number of photographed images of the target group stored in the auxiliary storage device 15 has reached a predetermined upper limit. If the selection unit 43 determines that the number of photographed images of the target group stored in the auxiliary storage device 15 has not reached the predetermined upper limit, the selection unit 43 stores newly acquired photographed images as photographed images of the target group in the auxiliary storage device 15. If the selection unit 43 determines that the number of photographed images of the target group stored in the auxiliary storage device 15 has reached the predetermined upper limit, the selection unit 43 determines whether to replace the photographed images of the target group stored in the auxiliary storage device 15 with the newly acquired photographed images. As a result, the number of photographed images stored in the auxiliary storage device 15 is limited so as not to exceed the upper limit, thereby reducing the amount of data stored in the auxiliary storage device 15.

[0058] [Effects of this embodiment] The effects of this embodiment will be described. (1) When the number of photographed images of the target group stored in the auxiliary storage device 15 has reached a predetermined upper limit, the sorting unit 43 determines whether to replace the photographed images of the target group stored in the auxiliary storage device 15 with newly acquired photographed images. This limits the number of photographed images of each group stored in the auxiliary storage device 15 so as not to exceed the upper limit, thereby reducing the amount of data stored in the auxiliary storage device 15. Furthermore, the counting unit 42 counts the number of photographed images classified into the target group and stores the counted number of photographed images in the auxiliary storage device 15. This makes it possible to obtain the distribution of the abnormality degrees of the loom 100 without saving all of the photographed images. Therefore, it is possible to obtain the distribution of the abnormality degrees of the loom 100 while reducing the amount of data stored in the auxiliary storage device 15.

[0059] (2) The classification unit 41 calculates the degree of abnormality of the loom 100 as a score. Based on the calculated score, the classification unit 41 classifies the captured image into one of a plurality of groups. With this configuration, the degree of abnormality of the loom 100 can be detected in more detail than when the degree of abnormality of the loom 100 is detected in stages such as S, A, B, C or excellent, good, fair, and unacceptable.

[0060] (3) The classification unit 41 calculates the score so that the higher the degree of abnormality in the loom 100, the lower the score. If the score of the newly acquired photographed image is lower than the score of the photographed image with the highest score among the photographed images of the target group stored in the auxiliary storage device 15, the selection unit 43 replaces the photographed image with the highest score among the photographed images of the target group stored in the auxiliary storage device 15 with the newly acquired photographed image.

[0061] According to this configuration, captured images with low scores, i.e., captured images showing a high degree of abnormality in the loom 100, are preferentially stored in the auxiliary storage device 15. Therefore, compared to a case where captured images with high scores, i.e., captured images showing a low degree of abnormality in the loom 100, are preferentially stored in the auxiliary storage device 15, it is easier to use the captured images when investigating the cause of the abnormality in the loom 100.

[0062] (4) The classification unit 41 further classifies the captured images into one of a plurality of groups based on the capture time of the captured images. This makes it possible to obtain the time transition of the distribution of the abnormality degree of the loom 100. The time transition of the distribution of the abnormality degree of the loom 100 is useful for understanding the state of the loom 100 when, for example, the operating conditions of the loom 100 are changed while the loom 100 is in operation or the installation environment of the loom 100 is changed.

[0063] (5) The inspection system 10 is equipped with a display unit 17. The display unit 17 displays a table H showing the number of captured images classified into each group. This makes it easier for the operator to grasp the distribution of the abnormality levels of the loom 100.

[0064] (6) The display unit 17 displays Table H in a heat map format according to the number of captured images classified into each group. This makes it easier for the operator to grasp the distribution of the abnormality levels of the loom 100.

[0065] (7) When the number of photographed images of the target group stored in the auxiliary storage device 15 reaches a predetermined upper limit, the selection unit 43 determines whether to replace the photographed images of the target group stored in the auxiliary storage device 15 with newly acquired photographed images without terminating the storage of the photographed images of the target group. This allows photographed images more suitable for storage to be stored in the auxiliary storage device 15.

[0066] [Example of change] The above embodiment can be modified as follows: The above embodiment and the following modifications can be combined with each other within the scope of technical compatibility.

[0067] In the above embodiment, the classification unit 41 detects the degree of abnormality of the loom 100 each time a new image is acquired from the camera 11. However, this is not limited to this. The classification unit 41 may detect the degree of abnormality of the loom 100 when the number of images acquired from the camera 11 reaches a predetermined number.

[0068] In the above embodiment, the control device 14 acquires the captured image from the camera 11 every time the camera 11 captures an image, but this is not limited to this. The control device 14 may acquire a predetermined number of captured images from the camera 11 at once when the camera 11 has captured an image a predetermined number of times.

[0069] The classification unit 41 may detect the degree of abnormality of the loom 100 in stages, such as S, A, B, C or excellent, good, fair, or unacceptable. In the above embodiment, the classification unit 41 classifies the captured images into one of a plurality of groups based on the degree of abnormality in the loom 100 and the capture time of the captured image. However, the present invention is not limited to this.

[0070] The classification unit 41 may classify the captured images into one of a plurality of groups based solely on the degree of abnormality in the loom 100. The classification unit 41 may classify the captured images into one of a plurality of groups based on items other than the degree of abnormality and the time of capture of the loom 100. Items other than the time of capture include, for example, the operating conditions of the loom 100 and the installation environment of the loom 100.

[0071] The score increments used by the classification unit 41 when classifying captured images into one of a plurality of groups are not limited to 0.01 and may be changed as appropriate. The score increments used by the classification unit 41 when classifying captured images into one of a plurality of groups do not have to be uniform. For example, among scores calculated in the range of 0 to 1, the score increments in a specific score range may be finer than the score increments in other score ranges.

[0072] The multiple groups based on the scores may be, for example, three groups as follows: The first group is a group of looms 100 whose scores are equal to or less than a first score. The second group is a group of looms 100 whose scores are equal to or greater than a second score, where the second score is greater than the first score. The third group is a group of looms 100 whose scores are greater than the first score but less than the second score.

[0073] The interval of the shooting time when the classification unit 41 classifies the captured images into one of the plurality of groups is not limited to one minute and may be changed as appropriate. The predetermined upper limit number is not limited to five and may be changed as appropriate.

[0074] The predetermined upper limit number may be different for each group. In the above embodiment, when the score of a newly acquired photographed image is lower than the score of the photographed image with the highest score among the photographed images of the target group stored in the auxiliary storage device 15, the selection unit 43 replaces the photographed image with the highest score among the photographed images of the target group stored in the auxiliary storage device 15 with the newly acquired photographed image, but this is not limited to this.

[0075] For example, if the score of the newly acquired photographed image is higher than the lowest-scoring photographed image among the photographed images of the target group stored in the auxiliary storage device 15, the selection unit 43 may replace the lowest-scoring photographed image among the photographed images of the target group stored in the auxiliary storage device 15 with the newly acquired photographed image. In this case, the photographed image with the higher score, i.e., the photographed image in which the degree of abnormality of the loom 100 is low, is preferentially stored in the auxiliary storage device 15.

[0076] The number of cameras 11 may be multiple. The storage unit is not limited to the auxiliary storage unit 15. The storage unit may be, for example, the main storage unit 14b of the control device 14, a storage unit provided in the machine base 101, or a cloud.

[0077] The sorting unit 41, the counting unit 42, and the selection unit 43 are not limited to being the control device 14 of the computer 10b. The sorting unit 41, the counting unit 42, and the selection unit 43 may be, for example, a control device that controls the loom 100.

[0078] The display unit 17 does not have to display the table H in the heat map format. In the above embodiment, the classification unit 41 detected the degree of abnormality of the loom 100 based on the warp thread T shown in the captured image, but the classification unit 41 may also detect the degree of abnormality of the loom 100 based on the weft thread or components of the loom 100 shown in the captured image.

[0079] The textile machine inspection system 10 may be applied to textile machines other than the loom 100, such as a spinning machine. [Note] The technical ideas that can be understood from the above-described embodiments and modifications will be described below.

[0080] <Appendix 1> a classification unit that detects a degree of abnormality of the textile machine based on newly acquired captured images and classifies the newly acquired captured images into one of a plurality of groups based on the detected degree of abnormality of the textile machine; a memory unit that stores the captured images for each group; a counting unit that counts the number of captured images classified into a target group, which is the group into which the newly acquired captured images have been classified, and stores the counted number of captured images in the memory unit; and a selection unit that determines whether or not the number of captured images of the target group stored in the memory unit has reached a predetermined upper limit number, and if it determines that the number of captured images of the target group stored in the memory unit has not reached the predetermined upper limit number, stores the newly acquired captured images in the memory unit as the captured images of the target group, and if it determines that the number of captured images of the target group stored in the memory unit has reached the predetermined upper limit number, determines whether or not to replace the captured image of the target group stored in the memory unit with the newly acquired captured image.

[0081] <Appendix 2> The textile machine inspection system described in Appendix 1, wherein the classification unit calculates the degree of abnormality of the textile machine as a score and classifies the captured image into one of the multiple groups based on the calculated score.

[0082] <Appendix 3> The classification unit calculates the score so that the higher the degree of abnormality in the textile machine, the lower the score, and the sorting unit replaces the image with the highest score among the images of the target group stored in the memory unit with the newly acquired image if the score of the newly acquired image is lower than the score of the image with the highest score among the images of the target group stored in the memory unit.

[0083] <Appendix 4> 4. The textile machine inspection system according to any one of claims 1 to 3, wherein the classification unit further classifies the captured image into one of the plurality of groups based on the time the captured image was captured.

[0084] <Appendix 5> 5. The textile machine inspection system according to any one of claims 1 to 4, further comprising a display unit, which displays a table showing the number of captured images classified into each group.

[0085] <Appendix 6> 6. The textile machine inspection system according to claim 5, wherein the display unit displays the table in a heat map format according to the number of captured images classified into each group. [Explanation of symbols]

[0086] 10...Textile machine inspection system, 11...camera, 15...auxiliary memory device as memory unit, 17...display unit, 41...classification unit, 42...counting unit, 43...sorting unit, P...captured image, H...table.

Claims

1. A textile machine inspection system that inspects a textile machine based on an image captured by a camera, a classification unit that detects a degree of abnormality of the textile machine based on the newly acquired photographed image, and classifies the newly acquired photographed image into one of a plurality of groups based on the detected degree of abnormality of the textile machine; a storage unit that stores the captured images for each group; a counting unit that counts the number of the photographed images that have been classified into a target group, which is the group into which the newly acquired photographed image has been classified, and stores the counted number of the photographed images in the storage unit; a selection unit that determines whether or not the number of the photographed images of the target group stored in the storage unit has reached a predetermined upper limit, and when it is determined that the number of the photographed images of the target group stored in the storage unit has not reached the predetermined upper limit, stores a newly acquired photographed image in the storage unit as the photographed image of the target group, and when it is determined that the number of the photographed images of the target group stored in the storage unit has reached the predetermined upper limit, determines whether or not to replace the photographed image of the target group stored in the storage unit with the newly acquired photographed image; A textile machine inspection system comprising:

2. The textile machine inspection system according to claim 1, wherein the classification unit calculates the degree of abnormality of the textile machine as a score, and classifies the captured image into one of the plurality of groups based on the calculated score.

3. the classification unit calculates the score so that the higher the degree of abnormality of the textile machine, the lower the score; 3. The textile machine inspection system of claim 2, wherein the sorting unit replaces the image with the highest score among the images of the target group stored in the memory unit with the newly acquired image if the score of the newly acquired image is lower than the score of the image with the highest score among the images of the target group stored in the memory unit.

4. The textile machine inspection system according to claim 1 , wherein the classification unit further classifies the captured image into one of the plurality of groups based on the time the captured image was captured.

5. A display unit is provided, The textile machine inspection system according to claim 1 , wherein the display unit displays a table indicating the number of the captured images classified into each group.

6. The textile machine inspection system according to claim 5 , wherein the display unit displays the table in a heat map format according to the number of the captured images classified into each group.

Images