A method, system and terminal for detecting the appearance of a pole end face

Abstract

The present invention relates to a method, system and terminal for detecting the appearance of a pole end face, and relates to the field of pole detection technology. The method comprises: positioning the pole based on a preset positioning device and collecting and acquiring illumination image information; retrieving annular light image information and spherical integral light image information based on the illumination image information; analyzing and determining defects such as stripes, scratches, incompleteness, concave spots, aluminum nail appearance, dirtiness, etc. according to the annular light image information and the spherical integral light image information and using them as appearance detection result information, and outputting the appearance detection result information. The present invention has the effect of improving the overall product quality in the pole production process.

Description

Technical Field

[0001] The present invention relates to the technical field of pole detection, and in particular to a pole end face appearance detection method, system and terminal. Background Art

[0002] The pole refers to the component that plays a key connecting and conductive role in equipment such as batteries and vacuum circuit breakers. In order to ensure that the poles in batteries, electrical equipment, etc. can be used normally, it is generally necessary to inspect and test the poles in terms of quality, performance, etc.

[0003] When inspecting the appearance of the pole, the inspector is generally required to wear the corresponding protective equipment, place the pole on a workbench with sufficient light and a clean background, and directly observe the overall appearance of the pole with the naked eye to see if there are obvious defects such as deformation, damage, cracks, etc., and check whether there are impurities such as oil, dust, rust, etc. on the surface of the pole. The inspector then uses auxiliary tools such as calipers, magnifying glasses or flashlights to measure key dimensions such as the length, diameter, and height of the pole, and conducts a more detailed observation of the gaps on the surface, bottom or side of the pole, so as to ensure the product quality during the production process of the pole.

[0004] Since the appearance inspection of the pole is generally carried out manually by inspectors, and different inspectors have different judgment standards for the appearance defects of the pole, it is easy to cause inconsistent product quality during the production process of the pole, and further lead to a decline in the overall product quality during the production process of the pole. Summary of the invention

[0005] In order to improve the overall product quality in the pole production process, the present invention provides a pole end face appearance detection method, system and terminal.

[0006] In a first aspect, the present invention provides a method for detecting the appearance of a pole end face, which adopts the following technical solution:

[0007] A method for detecting the appearance of a pole end face, comprising:

[0008] Based on the preset positioning device, the pole is positioned and the illumination image information is acquired;

[0009] Retrieve the annular light picture information and the spherical integral light picture information based on the illumination image information;

[0010] Determine the stripe defect information based on the annular light image information and the spherical integral light image information;

[0011] Retrieving stripe contrast based on stripe defect information;

[0012] When the stripe contrast is greater than the preset stripe reference contrast, the stripe length value and the stripe quantity value are retrieved based on the stripe defect information;

[0013] The appearance inspection result information is determined according to the stripe length value and the stripe quantity value, and the appearance inspection result information is output.

[0014] Optionally, the method for determining the appearance inspection result information includes:

[0015] Determine whether the stripe length value is greater than a preset stripe reference length value;

[0016] If yes, then determining the stripe length abnormality information corresponding to the stripe length value according to the correspondence between the stripe length value and the preset stripe length abnormality information, and using the stripe length abnormality information as the appearance detection result information;

[0017] If not, determine whether the stripe quantity value is greater than a preset stripe reference quantity value;

[0018] If yes, then determine the stripe quantity abnormality information according to the stripe quantity value and the stripe length value, and use the stripe quantity abnormality information as the appearance detection result information;

[0019] If not, the preset normal scratch detection information is output as the appearance detection result information.

[0020] Optionally, the method further includes steps after outputting the appearance inspection result information, which are as follows:

[0021] performing mean filtering according to the ring light image information to form mean filtering information;

[0022] Performing dynamic threshold segmentation of bright spots according to mean filter information to form bright spot segmentation information;

[0023] Retrieving the bright spot contrast based on the bright spot segmentation information;

[0024] When the bright spot contrast is greater than the preset bright spot reference contrast, the bright spot area value and the bright spot size value are retrieved based on the bright spot segmentation information;

[0025] Determine whether the bright spot area value is greater than a preset bright spot reference area value;

[0026] If yes, then determine the scratch detection abnormality information corresponding to the bright spot area value according to the correspondence between the bright spot area value and the preset scratch detection abnormality information, and add the scratch detection abnormality information to the appearance inspection result information to form new appearance inspection result information;

[0027] If yes, when the bright spot contrast is greater than the preset defect reference contrast, the defect detection result information is determined based on the comparison result between the bright spot size value and the preset defect reference size value, and the defect detection result information is added to the appearance detection result information to form new appearance detection result information.

[0028] Optionally, the method further includes a step after adding the scratch detection abnormality information to the appearance detection result information to form new appearance detection result information, or after adding the defect detection result information to the appearance detection result information to form new appearance detection result information, which is specifically as follows:

[0029] According to the ring light image information, grayscale binarization threshold segmentation is performed on the bright spots to form pit detection information;

[0030] Retrieving pit contrast and pit size values ​​based on pit detection information;

[0031] Determine whether the concave dot contrast is greater than a preset concave dot reference contrast and whether the concave dot size is greater than a preset concave dot reference size;

[0032] If both are greater than, determining the pit detection abnormality information according to the pit contrast and the pit size value, and adding the pit detection abnormality information to the appearance detection result information to form new appearance detection result information;

[0033] If they are not greater than , the yellow patch is segmented according to the annular light image information and the spherical integral light image information to form aluminum nail detection information;

[0034] Retrieve the aluminum nail detection threshold and the aluminum nail detection area value based on the aluminum nail detection information;

[0035] When the aluminum nail detection threshold is not within the preset aluminum nail threshold reference interval and the aluminum nail detection area value is greater than the preset aluminum nail reference area value, the aluminum nail detection abnormality information is determined based on the aluminum nail detection threshold and the aluminum nail detection area value, and the aluminum nail detection abnormality information is added to the appearance inspection result information to form new appearance inspection result information.

[0036] Optionally, the method further includes a step after adding the abnormal information of the pit detection to the appearance detection result information to form new appearance detection result information, or after adding the abnormal information of the aluminum nail detection to the appearance detection result information to form new appearance detection result information, which is specifically as follows:

[0037] Segment the dark spots according to the spherical integrated illumination image information to form dirt detection information;

[0038] Retrieving dirt detection contrast and dirt detection area value based on dirt detection information;

[0039] At the same time, it is determined whether the dirt detection contrast is greater than the preset dirt reference contrast and whether the dirt detection area value is greater than the preset dirt reference area value;

[0040] If both are greater than, then the dirt detection abnormality information is determined according to the dirt detection contrast and the dirt detection area value, and the dirt detection abnormality information is added to the appearance detection result information to form new appearance detection result information;

[0041] If they are not greater than at the same time, continue to output the appearance inspection result information.

[0042] Optionally, the method for obtaining the illumination image information includes:

[0043] Obtain image detection information when no illumination is performed;

[0044] Based on the preset pole features, the image detection information is identified to form pole image information;

[0045] Retrieve the pole image position point and pole area value based on the pole image information;

[0046] According to the correspondence between the pole area value and the preset pole distance value, a pole distance value corresponding to the pole area value is determined;

[0047] Determine the illumination control information according to the pole distance value and the pole image position point;

[0048] Based on the illumination control information, the preset illumination device is controlled to operate, and the pole image information is reacquired as the illumination image information.

[0049] Optionally, the method for determining the lighting control information includes:

[0050] Calculate the vector deviation between the pole image position point and the preset image reference position point and use it as the pole position vector distance value;

[0051] According to the correspondence between the pole position vector distance value and the preset coaxial light illumination adjustment information, the coaxial light illumination adjustment information corresponding to the pole position vector distance value is determined;

[0052] According to the correspondence between the pole distance value and the preset distance reference brightness value, the distance reference brightness value corresponding to the pole distance value is determined;

[0053] According to the correspondence between the pole position vector distance value and the preset position distance power influence value, the position distance power influence value corresponding to the pole position vector distance value is determined;

[0054] Calculate the sum of the distance reference brightness value and the position distance power influence value and use it as the power adjustment value;

[0055] Retrieve coaxial light illumination model information, annular light illumination model information and spherical integral light illumination model information based on preset illumination equipment;

[0056] Select and determine the lighting time control information according to the coaxial light lighting model information, the annular light lighting model information and the spherical integral light lighting model information;

[0057] Based on the coaxial light illumination adjustment information, the power adjustment value and the illumination time control information are combined to serve as the illumination control information.

[0058] Optionally, the method for determining the lighting time control information includes:

[0059] According to the correspondence between the coaxial light illumination model information and the preset coaxial light illumination reference time value, the coaxial light illumination reference time value corresponding to the coaxial light illumination model information is determined;

[0060] According to the correspondence between the ring light illumination model information and the preset ring light illumination reference time value, the ring light illumination reference time value corresponding to the ring light illumination model information is determined;

[0061] According to the correspondence between the spherical integral illumination illumination model information and the preset spherical integral illumination illumination reference time value, the spherical integral illumination illumination reference time value corresponding to the spherical integral illumination illumination model information is determined;

[0062] Based on the coaxial light illumination reference time value and the annular light illumination reference time value, a larger value is selected and used as the annular light operation time value;

[0063] Based on the coaxial light illumination reference time value and the spherical integral light illumination reference time value, a larger value is selected and used as the spherical integral light operation time value;

[0064] Based on the ring light running time value and the spherical integrated light running time value, a larger value is selected and used as the combined light running time value;

[0065] The lighting time control information is formed based on the ring light running time value, the spherical integrated light running time value and the combined light running time value.

[0066] In a second aspect, the present invention provides a pole end face appearance inspection system, which adopts the following technical solution:

[0067] A pole end face appearance detection system, comprising:

[0068] An acquisition module, used to acquire illumination image information and image detection information;

[0069] A memory, used to store a program of the pole end face appearance detection method as described in any one of the first aspects;

[0070] The processor loads and executes the program in the memory.

[0071] In a third aspect, the present invention provides an intelligent terminal, which adopts the following technical solution:

[0072] An intelligent terminal comprises a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executes a method for detecting the appearance of a pole end face as described in any one of the first aspects.

[0073] In summary, the present invention includes at least one of the following beneficial technical effects:

[0074] 1. Position the pole and collect the illumination image information to retrieve the annular light image information and the spherical integral light image information, then analyze and determine the stripe defect information and retrieve the stripe contrast. When it is greater than the preset stripe reference contrast, the appearance inspection result information is determined and output, so as to detect the abnormality of the appearance of the pole through the image, thereby ensuring the consistency of the inspection results and improving the overall product quality in the pole production process;

[0075] 2. By judging whether the stripe length value is greater than a preset stripe reference length value, and if it is greater, determining the stripe length abnormality information through the stripe length value query and using it as the appearance inspection result information; if it is not greater, determining whether the stripe quantity value is greater than a preset stripe reference quantity value, using the stripe quantity abnormality information or the scratch detection normal information as the appearance inspection result information, thereby improving the accuracy of the appearance inspection result information obtained;

[0076] 3. Determine the bright spot segmentation information and retrieve the bright spot contrast through the ring light image information. When the bright spot contrast is greater than the preset bright spot reference contrast, retrieve the bright spot area value and the bright spot size value through the bright spot segmentation information. Then, based on the judgment result of whether the bright spot area value is greater than the preset bright spot reference area value, update the appearance inspection result information, thereby improving the accuracy of the acquired appearance inspection result information. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 is a flow chart of a method for detecting the appearance of a pole end face according to an embodiment of the present application;

[0078] Figure 2 is a flow chart of a method for determining appearance inspection result information in an embodiment of the present application;

[0079] Figure 3 It is a method flow chart of the steps after outputting appearance inspection result information in an embodiment of the present application. DETAILED DESCRIPTION

[0080] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0081] A method for detecting the appearance of a pole end face, which controls the illumination after positioning the pole to obtain different annular light images and spherical integral light images, and then analyzes and determines whether there is any abnormality in the appearance of the pole, thereby ensuring the consistency of the detection results and improving the overall product quality in the pole production process.

[0082] Reference Figure 1 The embodiment of the present invention discloses a method for detecting the appearance of a pole end face, which comprises:

[0083] Step S100: positioning the pole based on a preset positioning device and acquiring illumination image information.

[0084] Among them, the positioning device refers to a device used to fix the pole, and the positioning device can be a caliper tool or other devices. The illumination image information refers to the image information collected after the pole is illuminated, and the illumination image information is detected and obtained by a camera preset at a position near the positioning device.

[0085] Step S101: Retrieve annular light picture information and spherical integral light picture information based on illumination image information.

[0086] The illumination image information includes annular light image information and spherical integral light image information. The annular light image information refers to the polar image information collected after irradiation with annular light and coaxial light, and the spherical integral light image information refers to the polar image information collected after irradiation with spherical integral light and coaxial light. The annular light image information and the spherical integral light image information are retrieved through the illumination image information for subsequent use.

[0087] Step S102: Analyze and determine the stripe defect information based on the annular light image information and the spherical integral light image information.

[0088] The stripe defect information refers to the detection information obtained by analyzing the stripe defects in the image. The stripe defects in the image are searched and detected by using the lines-gauss and derivate_gauss derivative operators, and the search and detection results are used as the stripe defect information for subsequent use.

[0089] Step S103: retrieve the stripe contrast based on the stripe defect information.

[0090] The stripe defect information includes stripe contrast, which refers to the contrast existing after the stripe defect is searched. The stripe contrast is retrieved through the stripe defect information to facilitate subsequent use.

[0091] Step S104: when the stripe contrast is greater than the preset stripe reference contrast, a stripe length value and a stripe quantity value are retrieved based on the stripe defect information.

[0092] The stripe reference contrast refers to the reference contrast corresponding to the absence of stripes, and the stripe reference contrast is obtained by pre-input. The stripe length value refers to the length value corresponding to the stripes in the image, and the stripe quantity value refers to the quantity value corresponding to the stripes in the image. The stripe defect information includes the stripe length value and the stripe quantity value.

[0093] When the stripe contrast is greater than the preset stripe reference contrast, it indicates that stripes exist at this time, so the stripe length value and the stripe quantity value are retrieved through the stripe defect information for subsequent use.

[0094] When the stripe contrast is not greater than the preset stripe reference contrast, it means that there is no stripe at this time, so the preset normal detection result information is directly output, and the normal detection result information is used as the appearance detection result information. The normal detection result information refers to the result information corresponding to the absence of stripes, and the normal detection result information is obtained after pre-input. In this embodiment, the stripe reference contrast is 20.

[0095] Step S105: determining appearance detection result information according to the stripe length value and the stripe quantity value, and outputting the appearance detection result information.

[0096] The appearance inspection result information refers to the result information after the appearance of the pole is inspected. The appearance inspection result information is determined by analyzing the stripe length value and the stripe quantity value, and the appearance inspection result information is output, so that the appearance of the pole is inspected according to the image inspection to ensure the consistency of the inspection results and improve the overall product quality in the production process of the pole. The specific steps for determining the appearance inspection result information refer to steps S200 to S204.

[0097] exist Figure 1 In step S105, in order to further ensure the rationality of the appearance inspection result information, it is necessary to further analyze and calculate the appearance inspection result information separately, specifically by Figure 2 The steps shown are explained in detail.

[0098] Reference Figure 2 , the method for determining appearance inspection result information includes the following steps:

[0099] Step S200: Determine whether the stripe length value is greater than a preset stripe reference length value. If yes, execute step S201; if no, execute step S202.

[0100] The stripe reference length value refers to the reference length value corresponding to the normal stripe, and the stripe reference length value is obtained by pre-input. By judging whether the stripe length value is greater than the preset stripe reference length value, it is judged whether the stripe length is abnormal. In this embodiment, the stripe reference length value is 5 mm.

[0101] Step S201: determining stripe length abnormality information corresponding to the stripe length value according to the correspondence between the stripe length value and the preset stripe length abnormality information, and using the stripe length abnormality information as appearance detection result information.

[0102] Among them, the stripe length abnormality information refers to the result information corresponding to the abnormal length of the stripes. Different stripe length values ​​correspond to different stripe length abnormality information. The stripe length abnormality information is obtained by querying from a database that stores different stripe length values ​​and corresponding stripe length abnormality information. The database is obtained after pre-input.

[0103] When the stripe length value is greater than the preset stripe reference length value, it indicates that the stripe length is abnormal. Therefore, the stripe length abnormality information is determined by querying the stripe length value, and the stripe length abnormality information is used as the appearance detection result information, thereby improving the accuracy of the obtained appearance detection result information.

[0104] Step S202: Determine whether the stripe quantity value is greater than a preset stripe reference quantity value. If yes, execute step S203; if no, execute step S204.

[0105] The stripe reference quantity value refers to the reference quantity value corresponding to the normal stripe, and the stripe reference quantity value is obtained by pre-input. When the stripe length value is not greater than the preset stripe reference length value, it means that the stripe length is not abnormal at this time, so by judging whether the stripe quantity value is greater than the preset stripe reference quantity value, it is judged whether the stripe quantity is abnormal. In this embodiment, the stripe reference quantity value is 2.

[0106] Step S203: determining the stripe quantity abnormality information according to the stripe quantity value and the stripe length value, and using the stripe quantity abnormality information as the appearance detection result information.

[0107] Among them, the abnormal number of stripes information refers to the result information corresponding to the abnormal number of stripes. The abnormal number of stripes information is determined by analyzing the stripe number value and the stripe length value, and the abnormal number of stripes information is used as the appearance detection result information, so as to facilitate subsequent use.

[0108] Further, in order to improve the accuracy of the acquired stripe quantity abnormal information, the method of obtaining the stripe quantity abnormal information includes:

[0109] Step S2031: Calculate the difference between the streak quantity value and the streak reference quantity value and use it as the streak deviation quantity value.

[0110] The stripe deviation quantity value refers to the deviation value when there is a deviation in the number of stripes. The difference between the stripe quantity value and the stripe reference quantity value is calculated and used as the stripe deviation quantity value for convenience in subsequent use.

[0111] Step S2031: Determine whether the stripe length value is less than the preset stripe multiple reference length value. If yes, execute step S2032; if no, execute step S2033.

[0112] The stripe-shaped multiple reference length value refers to the reference length value under normal circumstances when there are multiple stripes, and the stripe-shaped multiple reference length value is obtained after pre-input. By judging whether the stripe-shaped length value is greater than the preset stripe-shaped multiple reference length value, it is judged whether there are multiple stripes in an abnormal situation. In this embodiment, the stripe-shaped multiple reference length value is 3 mm.

[0113] Step S2032: outputting the preset stripe quantity reference information as stripe quantity abnormality information.

[0114] The stripe number reference information refers to the reference result information corresponding to the presence of multiple normal stripes, and the stripe number reference information is obtained after pre-input. When the stripe length value is less than the preset stripe multiple reference length value, it means that there is no abnormal situation of multiple stripes at this time, so the preset stripe number reference information is output and used as the stripe number abnormal information, so as to improve the accuracy of the acquired stripe number abnormal information.

[0115] Step S2033: Calculate the difference between the stripe length value and the stripe multiple reference length values ​​and use it as the stripe deviation length value.

[0116] The stripe deviation length value refers to the deviation value corresponding to the length deviation of multiple stripes. When the stripe length value is not less than the preset stripe multiple reference length value, it means that there are multiple stripes with abnormality. Therefore, the difference between the stripe length value and the stripe multiple reference length value is calculated and used as the stripe deviation length value for subsequent use.

[0117] Step S2034: determining the deviation quantity influence value corresponding to the stripe-shaped deviation quantity value according to the correspondence between the stripe-shaped deviation quantity value and the preset deviation quantity influence value.

[0118] The deviation quantity impact value refers to the impact degree value of the quantity deviation, and different stripe-shaped deviation quantity values ​​correspond to different deviation quantity impact values. The deviation quantity impact value is obtained by querying from a database storing different stripe-shaped deviation quantity values ​​and corresponding deviation quantity impact values, and the database is obtained after pre-input. The deviation quantity impact value is determined by querying the stripe-shaped deviation quantity value, which is convenient for subsequent use.

[0119] Step S2035: determining the deviation length influence value corresponding to the stripe deviation length value according to the correspondence between the stripe deviation length value and the preset deviation length influence value.

[0120] The deviation length impact value refers to the impact degree value of the length deviation. Different stripe-shaped deviation length values ​​correspond to different deviation length impact values. The deviation length impact value is obtained by querying from a database storing different stripe-shaped deviation length values ​​and corresponding deviation length impact values. The database is obtained after pre-input. The deviation length impact value is determined by querying the stripe-shaped deviation length value, which is convenient for subsequent use.

[0121] Step S2036: Calculate the sum of the deviation quantity impact value and the deviation length impact value and use it as the deviation comprehensive impact value.

[0122] The comprehensive impact value of deviation refers to the comprehensive impact value of length deviation and quantity deviation. The sum of the deviation quantity impact value and the deviation length impact value is calculated and used as the comprehensive impact value of deviation to facilitate subsequent use.

[0123] Step S2037: According to the correspondence between the deviation comprehensive influence value and the stripe comprehensive information, the stripe comprehensive information corresponding to the deviation comprehensive influence value is determined, and the stripe comprehensive information is used as the stripe quantity abnormality information.

[0124] The stripe-shaped comprehensive information refers to the stripe-shaped comprehensive result information indicating the length deviation and the quantity deviation. Different deviation comprehensive influence values ​​correspond to different stripe-shaped comprehensive information. The stripe-shaped comprehensive information is obtained by querying from a database storing different deviation comprehensive influence values ​​and corresponding stripe-shaped comprehensive information. The database is obtained after pre-input. The stripe-shaped comprehensive information is determined by querying the deviation comprehensive influence value, and the stripe-shaped comprehensive information is used as the stripe quantity abnormality information, thereby improving the accuracy of the acquired stripe quantity abnormality information.

[0125] Step S204: outputting preset normal scratch detection information as appearance detection result information.

[0126] The normal scratch detection information refers to the normal result information corresponding to the absence of scratches or other streaks, and the normal scratch detection information is obtained after pre-input. When the streak number value is not greater than the preset streak reference number value, it means that the number of streaks is not abnormal at this time, so the preset normal scratch detection information is output and used as the appearance inspection result information, thereby improving the accuracy of the obtained appearance inspection result information.

[0127] exist Figure 1 After step S105, in order to further ensure the rationality of the appearance inspection result information, it is necessary to further analyze and calculate the appearance inspection result information separately, specifically by Figure 3 The steps shown are explained in detail.

[0128] Reference Figure 3 , the steps after outputting the appearance inspection result information include the following steps:

[0129] Step S300: performing mean filtering according to the ring light image information to form mean filtering information.

[0130] The mean filter information refers to the detection information after the ring light image is mean filtered. The mean filter information is formed by performing mean filtering on the ring light image information using the mean_imge operator, which is convenient for subsequent use.

[0131] Step S301: performing dynamic threshold segmentation of bright spots according to mean filter information to form bright spot segmentation information.

[0132] The bright spot segmentation information refers to the detection information after the bright spots of the mean filter information are segmented. The bright spot segmentation information is obtained by segmenting the bright spots of the mean filter information using the dyn_threshold dynamic threshold, so as to facilitate subsequent use.

[0133] Step S302: Retrieve the bright spot contrast based on the bright spot segmentation information.

[0134] The bright spot segmentation information includes the bright spot contrast, which refers to the contrast of the bright spot. The bright spot contrast is retrieved through the bright spot segmentation information for easy subsequent use.

[0135] Step S303: when the bright spot contrast is greater than the preset bright spot reference contrast, the bright spot area value and the bright spot size value are retrieved based on the bright spot segmentation information.

[0136] The bright spot reference contrast refers to the reference contrast corresponding to the bright spot when there is no scratch, and the bright spot reference contrast is obtained by pre-input. The bright spot area value refers to the area value covered by the bright spot, and the bright spot size value refers to the size value corresponding to the range covered by the bright spot. The bright spot segmentation information includes the bright spot area value and the bright spot size value. In this embodiment, the bright spot reference contrast is 15.

[0137] When the bright spot contrast is greater than the preset bright spot reference contrast, it indicates that there is a possibility of scratches. Therefore, the bright spot area value and the bright spot size value are retrieved through the bright spot segmentation information for subsequent use.

[0138] When the bright spot contrast is not greater than the preset bright spot reference contrast, it indicates that there is no possibility of scratches at this time, so the preset normal detection result information is directly output and used as the appearance detection result information.

[0139] Step S304: Determine whether the bright spot area value is greater than a preset bright spot reference area value. If yes, execute step S305; if yes, execute step S306.

[0140] The bright spot reference area value refers to the reference area value corresponding to the absence of abnormal bright spots, and the bright spot reference area value is obtained after pre-input. By judging whether the bright spot area value is greater than the preset bright spot reference area value, it is judged whether there is a bright spot with an abnormal area. In this embodiment, the bright spot reference area value is 1 square millimeter.

[0141] Step S305: determining the scratch detection abnormality information corresponding to the bright spot area value according to the correspondence between the bright spot area value and the preset scratch detection abnormality information, and adding the scratch detection abnormality information to the appearance detection result information to form new appearance detection result information.

[0142] Among them, the scratch detection abnormal information refers to the abnormal result information corresponding to the existence of scratches, and different bright spot area values ​​correspond to different scratch detection abnormal information. The scratch detection abnormal information is obtained by querying from a database storing different bright spot area values ​​and corresponding scratch detection abnormal information, and the database is obtained after pre-input. When the bright spot area value is greater than the preset bright spot reference area value, it means that there is a bright spot with an abnormal area at this time, so the scratch detection abnormal information is determined by querying the bright spot area value, and the scratch detection abnormal information is added to the appearance detection result information to form new appearance detection result information, thereby improving the accuracy of the acquired appearance detection result information.

[0143] Step S306: When the bright spot contrast is greater than the preset defect reference contrast, the defect detection result information is determined according to the comparison result between the bright spot size value and the preset defect reference size value, and the defect detection result information is added to the appearance detection result information to form new appearance detection result information.

[0144] Among them, the defect reference contrast refers to the reference contrast corresponding to the presence of defect, and the defect reference contrast is obtained after pre-input. The defect reference size value refers to the reference size value corresponding to the bright spot when the defect exists, and the defect reference size value is obtained after pre-input. The defect detection result information refers to the result information of the defect. In this embodiment, the defect reference contrast is 30, and the defect reference size value is 0.5 mm.

[0145] When the bright spot area value is not greater than the preset bright spot reference area value, it means that there is no bright spot of abnormal area at this time. Therefore, when the bright spot contrast is not greater than the preset defect reference contrast, it means that there is no probability of defect at this time. Therefore, the preset normal detection result information is directly output, and the normal detection result information is used as the appearance detection result information.

[0146] When the bright spot contrast is greater than the preset defect reference contrast, it indicates that there is a probability of defect at this time. Therefore, the defect detection result information is determined by analyzing the comparison result between the bright spot size value and the preset defect reference size value, and the defect detection result information is added to the appearance detection result information to form new appearance detection result information, thereby improving the accuracy of the acquired appearance detection result information.

[0147] When the bright spot size value is greater than the defective reference size value, it indicates that there is a bright spot of defective size at this time, so the difference between the bright spot size value and the defective reference size value is calculated, and the preset defective abnormal size database is queried according to the calculated difference to determine the defective abnormal size result information, and the defective abnormal size result information is used as the defective detection result information. The defective abnormal size database is pre-input and stores different differences between the bright spot size value and the defective reference size value and the corresponding defective abnormal size result information. When the bright spot size value is not greater than the defective reference size value, it indicates that there is no bright spot of defective size at this time, so the incomplete normal size result information is output and used as the defective detection result information, and the incomplete normal size result information is obtained after pre-input.

[0148] exist Figure 3 After step S305 or step S306, in order to further ensure the rationality of the appearance inspection result information, it is necessary to perform further separate analysis and calculation on the appearance inspection result information, which is specifically described in detail through the following steps.

[0149] The step after adding the scratch detection abnormality information to the appearance detection result information to form new appearance detection result information, or after adding the defect detection result information to the appearance detection result information to form new appearance detection result information, includes the following steps:

[0150] Step S400: performing grayscale binarization threshold segmentation on the bright spots according to the ring light image information to form pit detection information.

[0151] The pit detection information refers to the detection information of the image used for pit detection. The pit detection information is obtained by segmenting the bright spots by using the threshold operator grayscale binarization threshold on the ring light image information, which is convenient for subsequent use.

[0152] Step S401: Retrieve the pit contrast and pit size values ​​based on the pit detection information.

[0153] The pit detection information includes pit contrast and pit size value, the pit contrast refers to the contrast corresponding to the pit detection, and the pit size value refers to the size value corresponding to the pit detection. The pit contrast and pit size value are retrieved through the pit detection information for subsequent use.

[0154] Step S402: determining whether the concave dot contrast is greater than a preset concave dot reference contrast and whether the concave dot size is greater than a preset concave dot reference size.

[0155] Among them, the concave point reference contrast refers to the reference contrast corresponding to the normal concave point, and the concave point reference size value refers to the reference size value corresponding to the normal concave point. The concave point reference contrast and the concave point reference size value are obtained after pre-input. By simultaneously judging whether the concave point contrast is greater than the preset concave point reference contrast and whether the concave point size value is greater than the preset concave point reference size value, it is judged whether there is a concave point abnormality. In this embodiment, the concave point reference contrast is 50, and the concave point reference size value is 0.5 mm.

[0156] Step S403: If both are greater than , determining the concave point detection abnormality information according to the concave point contrast and the concave point size value, and adding the concave point detection abnormality information to the appearance detection result information to form new appearance detection result information.

[0157] Among them, the abnormal information of concave point detection refers to the result information of the existence of abnormal concave points. If both are greater than , it means that there are abnormal concave points at this time. Therefore, the abnormal information of concave point detection is determined by analyzing the concave point contrast and the concave point size value, and the abnormal information of concave point detection is added to the appearance detection result information to form new appearance detection result information, thereby improving the accuracy of the acquired appearance detection result information.

[0158] The difference between the concave point contrast and the concave point reference contrast is calculated and used as the concave point deviation contrast, and the difference between the concave point size value and the concave point reference size value is calculated and used as the concave point deviation size value. The concave point detection database is queried through the concave point deviation contrast and the concave point deviation size value to determine the concave point detection abnormality information. The concave point detection database stores different concave point deviation contrasts and concave point deviation size values ​​and the corresponding different concave point detection abnormality information through pre-input.

[0159] Step S404: If they are not greater than , the yellow patch is segmented according to the annular light image information and the spherical integral light image information to form aluminum nail detection information.

[0160] Among them, the aluminum nail detection information refers to the detection information of the aluminum nails of the pole. If it is not greater than when at the same time, it means that there is no abnormal concave point at this time. Therefore, the aluminum nail detection information is obtained by using the trans_from_rgb operator to perform HSV color segmentation on the annular light image information and the spherical integral light image information to facilitate subsequent use.

[0161] Step S405: Retrieve the aluminum nail detection threshold and the aluminum nail detection area value based on the aluminum nail detection information.

[0162] Among them, the aluminum nail detection information includes the aluminum nail detection threshold and the aluminum nail detection area value. The aluminum nail detection threshold refers to the threshold corresponding to the aluminum nail of the detected pole, and the aluminum nail detection area value refers to the area value corresponding to the aluminum nail of the detected pole. The aluminum nail detection threshold and the aluminum nail detection area value are retrieved through the aluminum nail detection information to facilitate subsequent use.

[0163] Step S406: When the aluminum nail detection threshold is not within the preset aluminum nail threshold reference interval and the aluminum nail detection area value is greater than the preset aluminum nail reference area value, the aluminum nail detection abnormality information is determined according to the aluminum nail detection threshold and the aluminum nail detection area value, and the aluminum nail detection abnormality information is added to the appearance detection result information to form new appearance detection result information.

[0164] Among them, the aluminum nail threshold reference interval refers to the threshold reference interval corresponding to the aluminum nail of the pole under normal circumstances, and the aluminum nail reference area value refers to the reference area value corresponding to the aluminum nail of the pole under normal circumstances. The aluminum nail threshold reference interval and the aluminum nail reference area value are both obtained after pre-input. The aluminum nail detection abnormality information refers to the result information when the aluminum nail of the pole is abnormal. When the aluminum nail detection threshold is not located in the preset aluminum nail threshold reference interval and the aluminum nail detection area value is greater than the preset aluminum nail reference area value, it means that the aluminum nail of the pole is abnormal at this time. Therefore, by analyzing the aluminum nail detection threshold and the aluminum nail detection area value, the aluminum nail detection abnormality information is determined, and the aluminum nail detection abnormality information is added to the appearance detection result information to form a new appearance detection result information, thereby improving the accuracy of the acquired appearance detection result information. In this embodiment, the aluminum nail threshold reference interval is 20-50, and the aluminum nail reference area value is 3 square millimeters.

[0165] The difference between the aluminum nail detection threshold and the aluminum nail threshold reference interval is calculated and used as the aluminum nail threshold deviation value, the difference between the aluminum nail detection area value and the aluminum nail reference area value is calculated and used as the aluminum nail area deviation value, and the aluminum nail detection database is queried for the aluminum nail threshold deviation value and the aluminum nail area deviation value to determine the aluminum nail detection abnormality information. The aluminum nail detection database stores different aluminum nail threshold deviation values ​​and aluminum nail area deviation values ​​and the corresponding aluminum nail detection abnormality information through pre-input.

[0166] After step S403 or step S406, in order to further ensure the rationality of the appearance inspection result information, it is necessary to perform further separate analysis and calculation on the appearance inspection result information, which is specifically described in detail through the following steps.

[0167] The step after adding the pit detection abnormality information to the appearance detection result information to form new appearance detection result information, or after adding the aluminum nail detection abnormality information to the appearance detection result information to form new appearance detection result information includes the following steps:

[0168] Step S500: Segment dark spots according to the spherical integrated light image information to form dirt detection information.

[0169] The dirt detection information refers to the detection information for dirt detection, which is formed by using the threshold operator to perform grayscale binarization threshold segmentation on the spherical integral light image information to facilitate subsequent use.

[0170] Step S501: Retrieve dirt detection contrast and dirt detection area value based on dirt detection information.

[0171] The dirt detection information includes dirt detection contrast and dirt detection area value. The dirt detection contrast refers to the contrast corresponding to dirt detection, and the dirt detection area value refers to the area value corresponding to dirt detection. The dirt detection contrast and dirt detection area value are retrieved through the dirt detection information for subsequent use.

[0172] Step S502: simultaneously determining whether the dirt detection contrast is greater than a preset dirt reference contrast and whether the dirt detection area value is greater than a preset dirt reference area value.

[0173] The dirt reference contrast refers to the reference contrast corresponding to the absence of dirt, and the dirt reference area value refers to the reference area value corresponding to the absence of dirt. Both the dirt reference contrast and the dirt reference area value are obtained after pre-input. By simultaneously judging whether the dirt detection contrast is greater than the preset dirt reference contrast and whether the dirt detection area value is greater than the preset dirt reference area value, it is judged whether dirt exists. In this embodiment, the dirt reference contrast is 50, and the dirt reference area value is 5 square millimeters.

[0174] Step S503: If both are greater than, determining dirt detection abnormality information according to the dirt detection contrast and the dirt detection area value, and adding the dirt detection abnormality information to the appearance detection result information to form new appearance detection result information.

[0175] Among them, the abnormal dirt detection information refers to the result information corresponding to the presence of abnormal dirt. If both are greater than , it means that dirt exists at this time. Therefore, the dirt detection contrast and the dirt detection area value are analyzed to determine the abnormal dirt detection information, and the abnormal dirt detection information is added to the appearance detection result information to form new appearance detection result information, thereby improving the accuracy of the acquired appearance detection result information.

[0176] The difference between the dirt detection contrast and the dirt reference contrast is calculated and used as the dirt deviation contrast, the difference between the dirt detection area value and the dirt reference area value is calculated and used as the dirt deviation area value, and the dirt detection database is queried by the dirt deviation contrast and the dirt deviation area value to obtain the dirt detection abnormality information. The dirt detection database stores different dirt deviation contrasts and dirt deviation area values ​​and the corresponding dirt detection abnormality information through pre-input.

[0177] Step S504: If they are not greater than at the same time, continue to output appearance inspection result information.

[0178] Among them, if they are not greater than at the same time, it means that there is no dirt at this time, so the appearance inspection result information continues to be output to improve the accuracy of the obtained appearance inspection result information.

[0179] exist Figure 1 In step S100 shown, in order to further ensure the rationality of the illumination image information, it is necessary to perform further separate analysis and calculation on the illumination image information, which is specifically described in detail through the following steps.

[0180] The method for obtaining illumination image information comprises the following steps:

[0181] Step S600: Acquire image detection information when no illumination is performed.

[0182] The image detection information refers to image information detected when the pole is not illuminated, and the image detection information is acquired through detection by a camera.

[0183] Step S601: Identify image detection information based on preset pole features to form pole image information.

[0184] Among them, the pole feature refers to the shape, color and other features of the pole, and the pole feature is obtained after pre-input. The pole image information refers to the image information of the detected pole, and the pole image information is formed by identifying the pole feature of the image detection information, so as to facilitate subsequent use.

[0185] Step S602: Retrieve pole image position points and pole area values ​​based on pole image information.

[0186] The pole image information includes the pole image position point and the pole area value. The pole image position point refers to the position point corresponding to the pole in the image, and the pole area value refers to the area value corresponding to the pole in the image. The pole image position point and the pole area value are retrieved through the pole image information for subsequent use.

[0187] Step S603: determining a pole distance value corresponding to the pole area value according to a correspondence between the pole area value and a preset pole distance value.

[0188] The pole distance value refers to the distance between the pole and the position where the image is captured. Different pole area values ​​correspond to different pole distance values. The pole distance value is obtained by querying from a database storing different pole area values ​​and corresponding pole distance values. The pole distance value is determined by querying the pole area value to facilitate subsequent use.

[0189] Step S604: determining the illumination control information according to the pole distance value and the pole image position point.

[0190] The illumination control information refers to the control information used to control the operation of the illumination device, and the illumination device refers to the device used to illuminate the pole with coaxial light, annular light, and spherical integrated light. The illumination control information is determined by analyzing the pole distance value and the pole image position point, which is convenient for subsequent use. The specific method for determining the illumination control information refers to steps S700 to S707.

[0191] Step S605: controlling the preset lighting equipment to operate based on the lighting control information, and reacquiring the pole image information as the lighting image information.

[0192] The preset lighting equipment is controlled to operate through the lighting control information, and the pole image information is reacquired as the lighting image information, thereby improving the accuracy of the acquired lighting image information.

[0193] In step S604, in order to further ensure the rationality of the illumination control information, it is necessary to perform further separate analysis and calculation on the illumination control information, which is specifically described in detail through the following steps.

[0194] The method for determining the lighting control information comprises the following steps:

[0195] Step S700: Calculate the vector deviation between the pole image position point and the preset image reference position point and use it as the pole position vector distance value.

[0196] Among them, the image reference position point refers to the corresponding center position point in the detection image, and the image reference position point is obtained after pre-input. The pole position vector distance value refers to the vector distance value corresponding to the deviation between the pole position and the image center position. By calculating the vector deviation between the pole image position point and the preset image reference position point and using it as the pole position vector distance value, it is convenient for subsequent use.

[0197] Step S701: determining the coaxial light illumination adjustment information corresponding to the pole position vector distance value according to the correspondence between the pole position vector distance value and the preset coaxial light illumination adjustment information.

[0198] Among them, the coaxial light illumination adjustment information refers to the adjustment information that needs to be adjusted when controlling the illumination device to perform coaxial light. Different pole position vector distance values ​​correspond to different coaxial light illumination adjustment information. The coaxial light illumination adjustment information is obtained by querying from a database storing different pole position vector distance values ​​and corresponding coaxial light illumination adjustment information. The database is obtained after pre-input. The coaxial light illumination adjustment information is determined by querying the pole position vector distance value to facilitate subsequent use.

[0199] Step S702: determining a distance reference brightness value corresponding to the pole distance value according to a correspondence between the pole distance value and a preset distance reference brightness value.

[0200] The distance reference brightness value refers to the reference brightness value output according to the distance. Different pole distance values ​​correspond to different distance reference brightness values. The distance reference brightness value is obtained by querying from a database storing different pole distance values ​​and corresponding distance reference brightness values. The database is obtained after pre-input. The distance reference brightness value is determined by querying the pole distance value to facilitate subsequent use.

[0201] Step S703: determining the position distance power impact value corresponding to the pole position vector distance value according to the corresponding relationship between the pole position vector distance value and the preset position distance power impact value.

[0202] Among them, the position distance power impact value refers to the impact value corresponding to the impact of the deviation distance of the pole position on the power. Different pole position vector distance values ​​correspond to different position distance power impact values. The position distance power impact value is obtained by querying from a database storing different pole position vector distance values ​​and corresponding position distance power impact values. The database is obtained after pre-input. The position distance power impact value is determined by querying the pole position vector distance value to facilitate subsequent use.

[0203] Step S704: Calculate the sum of the distance reference brightness value and the position distance power influence value and use it as the power adjustment value.

[0204] The power adjustment value refers to the brightness value after the power is adjusted. The sum of the distance reference brightness value and the position distance power influence value is calculated and used as the power adjustment value for subsequent use.

[0205] Step S705: Retrieving coaxial light illumination model information, annular light illumination model information and spherical integral light illumination model information based on a preset illumination device.

[0206] Among them, the coaxial light lighting model information refers to the model information of the lighting components for coaxial light lighting in the lighting equipment, the ring light lighting model information refers to the model information of the lighting components for ring light lighting in the lighting equipment, and the spherical integral light lighting model information refers to the model information of the lighting components for spherical integral light lighting in the lighting equipment. By retrieving the coaxial light lighting model information, the ring light lighting model information and the spherical integral light lighting model information from the model of the lighting equipment, subsequent use is facilitated.

[0207] Step S706: Select and determine the illumination time control information according to the coaxial light illumination model information, the annular light illumination model information and the spherical integral light illumination model information.

[0208] The lighting time control information refers to the time control information used to control the duration of the lighting equipment for coaxial lighting, annular lighting and spherical integral lighting. The lighting time control information is determined by analyzing and selecting the coaxial lighting model information, annular lighting model information and spherical integral lighting model information, so as to facilitate subsequent use.

[0209] Step S707: combining the coaxial light illumination adjustment information, the power adjustment value and the illumination time control information to serve as illumination control information.

[0210] Among them, by combining the coaxial light illumination adjustment information, the power adjustment value and the illumination time control information as the illumination control information, the accuracy of the obtained illumination control information is improved, and it is convenient to control the illumination equipment to adjust the position with the coaxial light illumination adjustment information, and then control the illumination equipment to perform coaxial light illumination, annular light illumination and spherical integrated light illumination respectively according to the power adjustment value and the illumination time control information, so as to facilitate the subsequent pole detection.

[0211] In step S706, in order to further ensure the rationality of the lighting time control information, it is necessary to perform further separate analysis and calculation on the lighting time control information, which is specifically described in detail through the following steps.

[0212] The method for determining lighting time control information comprises the following steps:

[0213] Step S800: determining a coaxial light illumination reference time value corresponding to the coaxial light illumination model information according to a correspondence between the coaxial light illumination model information and a preset coaxial light illumination reference time value.

[0214] Among them, the coaxial light illumination reference time value refers to the reference time value that needs to be run when coaxial light illumination is performed. Different coaxial light illumination model information corresponds to different coaxial light illumination reference time values. The coaxial light illumination reference time value is obtained by querying from a database storing different coaxial light illumination model information and corresponding coaxial light illumination reference time values. The database is obtained after pre-input. The coaxial light illumination reference time value is determined by querying the coaxial light illumination model information, which is convenient for subsequent use.

[0215] Step S801: determining a ring light illumination reference time value corresponding to the ring light illumination model information according to the correspondence between the ring light illumination model information and the preset ring light illumination reference time value.

[0216] The reference time value of the ring light illumination refers to the reference time value required to run when the ring light illumination is performed. Different ring light illumination model information corresponds to different reference time values ​​of the ring light illumination. The reference time value of the ring light illumination is obtained by querying from a database storing different ring light illumination model information and corresponding reference time values ​​of the ring light illumination. The database is obtained after pre-input. The reference time value of the ring light illumination is determined by querying the ring light illumination model information, which is convenient for subsequent use.

[0217] Step S802: determining the spherical integral lighting reference time value corresponding to the spherical integral lighting model information according to the correspondence between the spherical integral lighting model information and the preset spherical integral lighting reference time value.

[0218] Among them, the spherical integral illumination reference time value refers to the reference time value that needs to be run when performing spherical integral illumination. Different spherical integral illumination model information corresponds to different spherical integral illumination reference time values. The spherical integral illumination reference time value is obtained by querying from a database storing different spherical integral illumination model information and corresponding spherical integral illumination reference time values. The database is obtained after pre-input. The spherical integral illumination reference time value is determined by querying the spherical integral illumination model information, which is convenient for subsequent use.

[0219] Step S803: selecting a larger value based on the coaxial light illumination reference time value and the ring light illumination reference time value and using the larger value as the ring light operation time value.

[0220] Among them, the ring light running time value refers to the time value required to run when performing ring light illumination and collecting images. By comparing the coaxial light illumination reference time value with the ring light illumination reference time value, and selecting the larger value as the ring light running time value, it is convenient for subsequent use.

[0221] Step S804: based on the coaxial light illumination reference time value and the spherical integrated light illumination reference time value, a larger value is selected as the spherical integrated light operation time value.

[0222] Among them, the spherical integral illumination running time value refers to the time value required to run when performing spherical integral illumination illumination and collecting images. By comparing the coaxial light illumination benchmark time value with the spherical integral illumination benchmark time value, and selecting the larger value as the spherical integral illumination running time value, it is convenient for subsequent use.

[0223] Step S805: selecting a larger value based on the annular light running time value and the spherical integrated light running time value and using the larger value as the combined light running time value.

[0224] Among them, the combined illumination running time value refers to the time value required to run when performing ring light illumination and spherical integral illumination illumination and collecting images. By comparing the ring light running time value with the spherical integral illumination running time value, and selecting the larger value as the combined illumination running time value, it is convenient for subsequent use.

[0225] Step S806: Combining the annular light running time value, the spherical integrated light running time value and the combined light running time value to form lighting time control information.

[0226] Among them, by combining the ring light running time value, the spherical integrated light running time value and the combined light running time value, the lighting time control information is formed, thereby improving the accuracy of the obtained lighting time control information.

[0227] Based on the same inventive concept, an embodiment of the present invention provides a pole end face appearance detection system, comprising:

[0228] An acquisition module, used to acquire illumination image information and image detection information;

[0229] A memory, used to store a program of the pole end face appearance detection method as described above;

[0230] The processor loads and executes the program in the memory.

[0231] Based on the same inventive concept, an embodiment of the present invention provides an intelligent terminal, including a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute the above-mentioned pole end face appearance detection method.

[0232] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0233] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A method for detecting the appearance of a pole end face, characterized in that: include: Based on the preset positioning device, the pole is positioned and the illumination image information is acquired; Retrieve the annular light picture information and the spherical integral light picture information based on the illumination image information; Determine the stripe defect information based on the annular light image information and the spherical integral light image information; Retrieving stripe contrast based on stripe defect information; When the stripe contrast is greater than the preset stripe reference contrast, the stripe length value and the stripe quantity value are retrieved based on the stripe defect information; Determine appearance inspection result information according to the stripe length value and the stripe quantity value, and output the appearance inspection result information; The method for obtaining illumination image information includes: Obtain image detection information when no illumination is performed; Based on the preset pole features, the image detection information is identified to form pole image information; Retrieve the pole image position point and pole area value based on the pole image information; According to the correspondence between the pole area value and the preset pole distance value, a pole distance value corresponding to the pole area value is determined; Determine the illumination control information according to the pole distance value and the pole image position point; Based on the illumination control information, the preset illumination device is controlled to operate, and the pole image information is reacquired as the illumination image information.

2. The pole end face appearance detection method according to claim 1, characterized in that: Methods for determining appearance inspection result information include: Determine whether the stripe length value is greater than a preset stripe reference length value; If yes, then determining the stripe length abnormality information corresponding to the stripe length value according to the correspondence between the stripe length value and the preset stripe length abnormality information, and using the stripe length abnormality information as the appearance detection result information; If not, determine whether the stripe quantity value is greater than a preset stripe reference quantity value; If yes, then determine the stripe quantity abnormality information according to the stripe quantity value and the stripe length value, and use the stripe quantity abnormality information as the appearance detection result information; If not, the preset normal scratch detection information is output as the appearance detection result information.

3. The pole end face appearance detection method according to claim 2, characterized in that: The steps after outputting the appearance inspection result information are also included, which are as follows: Performing mean filtering according to the ring light image information to form mean filtering information; Performing dynamic threshold segmentation of bright spots according to mean filter information to form bright spot segmentation information; Retrieving the bright spot contrast based on the bright spot segmentation information; When the bright spot contrast is greater than the preset bright spot reference contrast, the bright spot area value and the bright spot size value are retrieved based on the bright spot segmentation information; Determine whether the bright spot area value is greater than a preset bright spot reference area value; If yes, then determine the scratch detection abnormality information corresponding to the bright spot area value according to the correspondence between the bright spot area value and the preset scratch detection abnormality information, and add the scratch detection abnormality information to the appearance inspection result information to form new appearance inspection result information; If yes, when the bright spot contrast is greater than the preset defect reference contrast, the defect detection result information is determined based on the comparison result between the bright spot size value and the preset defect reference size value, and the defect detection result information is added to the appearance detection result information to form new appearance detection result information.

4. The pole end face appearance detection method according to claim 3, characterized in that: The method further includes a step after adding the scratch detection abnormality information to the appearance detection result information to form new appearance detection result information, or after adding the defect detection result information to the appearance detection result information to form new appearance detection result information, which is specifically as follows: According to the ring light image information, grayscale binarization threshold segmentation is performed on the bright spots to form pit detection information; Retrieving pit contrast and pit size values ​​based on pit detection information; Determine whether the concave dot contrast is greater than a preset concave dot reference contrast and whether the concave dot size is greater than a preset concave dot reference size; If both are greater than, determining the pit detection abnormality information according to the pit contrast and the pit size value, and adding the pit detection abnormality information to the appearance detection result information to form new appearance detection result information; If they are not greater than , the yellow patch is segmented according to the annular light image information and the spherical integral light image information to form aluminum nail detection information; Retrieve the aluminum nail detection threshold and the aluminum nail detection area value based on the aluminum nail detection information; When the aluminum nail detection threshold is not within the preset aluminum nail threshold reference interval and the aluminum nail detection area value is greater than the preset aluminum nail reference area value, the aluminum nail detection abnormality information is determined based on the aluminum nail detection threshold and the aluminum nail detection area value, and the aluminum nail detection abnormality information is added to the appearance inspection result information to form new appearance inspection result information.

5. The pole end face appearance detection method according to claim 4, characterized in that: The method further includes a step after adding the pit detection abnormality information to the appearance detection result information to form new appearance detection result information, or after adding the aluminum nail detection abnormality information to the appearance detection result information to form new appearance detection result information, which is specifically as follows: Segment the dark spots according to the spherical integrated illumination image information to form dirt detection information; Retrieving dirt detection contrast and dirt detection area value based on dirt detection information; At the same time, it is determined whether the dirt detection contrast is greater than the preset dirt reference contrast and whether the dirt detection area value is greater than the preset dirt reference area value; If both are greater than, then the dirt detection abnormality information is determined according to the dirt detection contrast and the dirt detection area value, and the dirt detection abnormality information is added to the appearance detection result information to form new appearance detection result information; If they are not greater than at the same time, continue to output the appearance inspection result information.

6. The pole end face appearance detection method according to claim 1, characterized in that: The method for determining the lighting control information includes: Calculate the vector deviation between the pole image position point and the preset image reference position point and use it as the pole position vector distance value; According to the correspondence between the pole position vector distance value and the preset coaxial light illumination adjustment information, the coaxial light illumination adjustment information corresponding to the pole position vector distance value is determined; According to the correspondence between the pole distance value and the preset distance reference brightness value, the distance reference brightness value corresponding to the pole distance value is determined; According to the correspondence between the pole position vector distance value and the preset position distance power influence value, the position distance power influence value corresponding to the pole position vector distance value is determined; Calculate the sum of the distance reference brightness value and the position distance power influence value and use it as the power adjustment value; Retrieve coaxial light illumination model information, annular light illumination model information and spherical integral light illumination model information based on preset illumination equipment; Select and determine the lighting time control information according to the coaxial light lighting model information, the annular light lighting model information and the spherical integral light lighting model information; Based on the coaxial light illumination adjustment information, the power adjustment value and the illumination time control information are combined to serve as the illumination control information.

7. The method for detecting the appearance of the pole end face according to claim 6, characterized in that: The method for determining the lighting time control information includes: According to the correspondence between the coaxial light illumination model information and the preset coaxial light illumination reference time value, the coaxial light illumination reference time value corresponding to the coaxial light illumination model information is determined; According to the correspondence between the ring light illumination model information and the preset ring light illumination reference time value, the ring light illumination reference time value corresponding to the ring light illumination model information is determined; According to the correspondence between the spherical integral illumination illumination model information and the preset spherical integral illumination illumination reference time value, the spherical integral illumination illumination reference time value corresponding to the spherical integral illumination illumination model information is determined; Based on the coaxial light illumination reference time value and the annular light illumination reference time value, a larger value is selected and used as the annular light operation time value; Based on the coaxial light illumination reference time value and the spherical integral light illumination reference time value, a larger value is selected and used as the spherical integral light operation time value; Based on the ring light running time value and the spherical integrated light running time value, a larger value is selected and used as the combined light running time value; The lighting time control information is formed based on the ring light running time value, the spherical integrated light running time value and the combined light running time value.

8. A pole end face appearance detection system, characterized in that: include: An acquisition module, used to acquire illumination image information and image detection information; A memory for storing a program of the pole end face appearance detection method according to any one of claims 1 to 7; The processor loads and executes the program in the memory.

9. An intelligent terminal, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executes the method for detecting the appearance of a pole end face according to any one of claims 1 to 7.

Images