Search

CN-121994133-A - Full-automatic intelligent detection system and method for appearance and size of shaft lever

CN121994133ACN 121994133 ACN121994133 ACN 121994133ACN-121994133-A

Abstract

The invention relates to the field of machine vision and automatic detection, and particularly discloses a full-automatic intelligent detection system and method for appearance and size of a shaft lever. The multi-station visual detection module comprises an area array camera and an independent line scanning camera which are arranged at two ends of the shaft rod, a coordinate system is realized through a high-precision calibration plate, the data processing and control module is integrated with a deep learning model, and the whole system realizes flow control and signal interaction through a PLC. The method comprises the steps of clamping a shaft rod, performing first image acquisition to calculate the total length and perform primary screening, rotating the shaft rod after the primary screening is qualified, performing second multi-angle image acquisition to calculate the diameter, roundness and runout values, synchronously detecting appearance defects based on line scanning images, and finally performing grading discharge on the shaft rod according to a comprehensive result and storing traceable data.

Inventors

  • YANG FANG

Assignees

  • 苏州未晞精密测控技术有限公司

Dates

Publication Date
20260508
Application Date
20260225

Claims (8)

  1. 1. A full-automatic intelligent detection system of axostylus axostyle outward appearance and size, its characterized in that includes: the multi-station visual detection module at least comprises a first visual detection unit and a second visual detection unit, and is used for respectively carrying out image acquisition on different parts of the shaft lever; The high-precision positioning and rotating mechanism is used for clamping and rotating the shaft lever; The data processing and control module is used for executing image processing, size calculation, roundness and run-out analysis and appearance defect identification; The intelligent sorting mechanism is used for carrying out grading discharge on the shaft rod according to the detection result; The user operation interface is used for parameter setting, image display and detection result display; the data processing and control module is in communication connection with the programmable logic controller to receive control instructions and feed back detection results.
  2. 2. The system of claim 1, wherein the multi-station visual inspection module comprises: The high-resolution area array camera is used for acquiring images of the end face and the diameter size of the shaft lever; A line scan camera for acquiring an expanded image of a cylindrical surface of the shaft as it rotates; and the high-precision calibration plate is used for calibrating the coordinate system of the area array camera to establish a unified measurement coordinate system.
  3. 3. The system of claim 1, wherein the data processing and control module is integrated with a deep learning based defect recognition model for recognizing whether there is at least one defect in the shaft surface deployment image that is missing a punch, missing a hole, or a punch anomaly.
  4. 4. A method for fully automating the detection of the appearance and dimensions of a shaft based on the system of any one of claims 1 to 3, comprising the steps of: Clamping the shaft lever and triggering detection; performing first image acquisition, calculating the total length of the shaft lever and obtaining a first detection result; if the first detection result is qualified, rotating the shaft rod and carrying out second image acquisition; calculating the diameter, roundness and runout values of the appointed position of the shaft lever based on the multi-angle image acquired for the second time; performing appearance defect detection based on the developed image of the surface of the shaft rod acquired by the line scanning camera; And carrying out grading discharge on the shaft rod according to the comprehensive detection results of the dimension, the form and position tolerance and the appearance defects.
  5. 5. The method of claim 4, wherein calculating the overall length of the shaft comprises: respectively acquiring edge images of the end faces of the two ends; converting the edge coordinates of two ends into the same world coordinate system through the coordinate mapping relation established by the high-precision calibration plate; The distance between the two end edges is calculated as the total length of the shaft.
  6. 6. The method of claim 4, wherein calculating roundness and runout values comprises: acquiring an image sequence under the condition that the shaft lever rotates at a plurality of angles, wherein the image sequence at least comprises 50 images covering 360-degree circumferences; Extracting bearing position edge points from each image and calculating the instantaneous diameter; And calculating a roundness value and a runout value through a roundness algorithm and a runout algorithm respectively based on the diameter data sequence.
  7. 7. The method of claim 4, wherein the performing appearance defect detection comprises: Inputting the developed image of the surface of the shaft rod acquired by the line scanning camera into a pre-trained convolutional neural network model; and obtaining a defect identification result output by the model.
  8. 8. The method as recited in claim 4, further comprising: storing detection data, images and judging results, and associating with the product identification; And supporting detection history inquiry and traceability based on product identification.

Description

Full-automatic intelligent detection system and method for appearance and size of shaft lever Technical Field The invention belongs to the field of machine vision and automatic detection, and particularly relates to a full-automatic intelligent detection system and method for appearance and size of a shaft lever. Background The shaft lever is used as a core part for transmitting power and motion in mechanical equipment, and the dimensional accuracy, the form and position tolerance and the surface quality of the shaft lever directly influence the assembly performance, the running stability and the service life of the equipment. In the field of precision manufacturing, the detection requirement on shaft lever workpieces enters the micron-sized category, and particularly, the critical dimensions and shape and position parameters such as total length, bearing position diameter, roundness, runout and the like must be strictly controlled within a tolerance zone. In addition, surface and structure defects such as missing and punching ribs, hole missing processing, abnormal punching ribs and the like possibly occurring in the processing process of the shaft lever are also important factors influencing the functional reliability of the shaft lever. Currently, in industrial production, detection of a shaft rod mainly depends on offline spot check by manually using a micrometer, a caliper, a roundness measuring instrument and the like. The method is low in efficiency, high in labor intensity, easy to be influenced by experience and fatigue degree of operators, large in subjective error and poor in consistency, and 100% full detection is difficult to realize. Although some automatic visual detection devices are applied to production lines, the functions of the automatic visual detection devices are single, namely, the automatic visual detection devices can only perform two-dimensional measurement of the end face size, cannot realize high-precision overall length calculation and form and position tolerance analysis of a cross-view field, can only perform simple identification of surface flaws, and cannot cope with stable detection of complex defects such as missing and punching ribs and the like which are related to a strong processing technology. More critical, the existing scheme generally regards dimension detection and appearance detection as independent procedures, and the dimension detection and the appearance detection are completed by different equipment step by step, so that the detection flow is long, the equipment layout is complex, the data cannot be uniformly managed and traced, and the comprehensive requirements of a modern intelligent production line on high efficiency, high precision, full automation and data integration are difficult to meet. Disclosure of Invention In order to solve the technical problems, the invention provides a full-automatic intelligent detection system and method for the appearance and the size of a shaft rod, which are used for solving the problems that in the prior art, the manual detection efficiency is low, the precision is poor, the detection is easy to miss, the automatic detection system has single function, the high-precision size measurement and the intelligent identification of complex appearance defects are difficult to consider, and the full-flow closed-loop control and the data tracing cannot be realized. The invention provides a full-automatic intelligent detection system for appearance and size of a shaft lever, which comprises the following components: the multi-station visual detection module at least comprises a first visual detection unit and a second visual detection unit, and is used for respectively carrying out image acquisition on different parts of the shaft lever; The high-precision positioning and rotating mechanism is used for clamping and rotating the shaft lever; The data processing and control module is used for executing image processing, size calculation, roundness and run-out analysis and appearance defect identification; The intelligent sorting mechanism is used for carrying out grading discharge on the shaft rod according to the detection result; The user operation interface is used for parameter setting, image display and detection result display; the data processing and control module is in communication connection with the programmable logic controller to receive control instructions and feed back detection results. Preferably, the multi-station vision detection module includes: The high-resolution area array camera is used for acquiring images of the end face and the diameter size of the shaft lever; A line scan camera for acquiring an expanded image of a cylindrical surface of the shaft as it rotates; and the high-precision calibration plate is used for calibrating the coordinate system of the area array camera to establish a unified measurement coordinate system. Preferably, the data processing and control module is integrated with a defec