Precision forging connecting rod surface quality automatic inspection equipment

Abstract

The invention provides automatic surface quality inspection equipment for a precisely forged connecting rod, relates to the technical fields of image acquisition and processing, algorithm theories, mode recognition, state estimation and the like, and mainly solves the problems of low inspection efficiency, high labor investment, easiness in visual fatigue, high false detection and omission ratio, low finished product percent of pass and the like of a manual visual inspection method. The automatic feeding and posture adjusting unit correctly feeds a connecting rod into and withdraws a station, the shadowless illumination and image acquisition unit acquires a surface image of the connecting rod, and the image processing module processes the surface image of the connecting rod and identifies defects. The grading unit divides the connecting rods into a qualified type and an unqualified type according to the surface quality detection result of the connecting rods; and the computer and control unit controls the automatic and continuous work of each unit. The method is mainly applied to inspection of the surface quality of the precise forging production line of the internal combustion engine connecting rod.

Description

Technical Field

[0001] The present invention relates to the technical fields of image acquisition (photographic technology), image processing, algorithm theory, pattern recognition and state estimation, and in particular to technology and equipment for automatic inspection of the surface quality of precision forged connecting rods. Background Art

[0002] Precision forging technology is widely used in the manufacturing of connecting rods for internal combustion engines. For small and medium-sized enterprises alone, the daily output of precision forged connecting rods has reached 10,000 pieces. The inspection of the surface quality of precision forged connecting rods is mainly carried out by manual visual inspection. The manual visual inspection method has problems such as low inspection efficiency, high labor input, easy visual fatigue, high false detection and missed detection rate, and low finished product qualification rate, which restricts the improvement of enterprise production efficiency and economic benefits. Summary of the invention

[0003] The purpose of the present invention is to provide an automatic surface quality inspection device for precision forged connecting rods with advanced technology, reasonable design, high inspection efficiency, low false detection and missed detection rate, and convenient use, in view of the defects and shortcomings of the current manual surface quality inspection method for precision forged connecting rods. It mainly consists of a hardware system including an automatic feeding unit, a posture adjustment unit, a shadowless lighting unit, an image acquisition unit, a grading unit, a computer and a control unit, and a software system including an image processing module, a model training module, a defect recognition algorithm module, a PLC ladder diagram control module, and a human-computer interaction module.

[0004] After the present invention adopts the above structure, its beneficial effects are: compared with manual inspection, with the close cooperation of various units of the system, the inspection efficiency is significantly improved; through shadowless lighting and image acquisition, the problem of visual fatigue is overcome; the surface defect pattern database, connecting rod surface defect recognition algorithm and connecting rod identification algorithm are applied to achieve accurate recognition of surface defects, eliminating problems such as false detection and missed detection; it is of great help in improving the comprehensive benefits of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present application, but do not constitute an improper limitation of the present invention. In the drawings:

[0006] Figure 1 It is a schematic diagram of the hardware system composition of the present invention;

[0007] Figure 2 It is a schematic diagram of the composition of the software system of the present invention.

[0008] Description of reference numerals:

[0009] 1-Hardware system, 11-Automatic feeding unit, 12-Posture adjustment unit, 13-Shadowless lighting unit, 14-Image acquisition unit, 15-Grading unit, 16-Computer and control unit, 2-Software system, 21-Image processing module, 22-Model training module, 23-Defect recognition algorithm module, 24-PLC ladder diagram control module, 25-Human-computer interaction module. DETAILED DESCRIPTION

[0010] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments, wherein the illustrative embodiments and descriptions are only used to explain the present invention but are not intended to limit the present invention.

[0011] like Figure 1As shown, the hardware system 1 of the automatic inspection equipment for the surface quality of precision forged connecting rods described in this specific embodiment mainly includes an automatic feeding unit 11, a posture adjustment unit 12, a shadowless lighting unit 13, an image acquisition unit 14, a connecting rod grading unit 15, and a computer and control unit 16. The automatic feeding unit feeds the connecting rods into the conveyor belt in an orderly manner; the posture adjustment unit performs posture adjustments such as flipping the connecting rods to ensure that the camera can collect images of the front and back planes and the sides of the connecting rods in a 360° direction; the shadowless lighting unit performs shadowless lighting on the front and back planes and the sides of the connecting rods in a 360° direction to ensure that the camera collects images that accurately reflect the surface quality status and defect characteristics; the image acquisition unit performs image acquisition on the front and back planes and the sides of the connecting rods in a 360° direction; the grading unit classifies the connecting rods into qualified and unqualified categories according to the surface quality inspection results of the connecting rods; the computer and control unit control the workpiece feeding, image acquisition, posture adjustment, grading and other units to automatically and continuously work through photoelectric switches, PLCs, motors, push rods and other devices. The software system 2 mainly includes an image processing module 21, a model training module 22, a defect recognition algorithm module 23, a PLC ladder diagram control module 24 and a human-computer interaction module 25. The image processing module pre-processes the collected connecting rod surface image, performs denoising, threshold segmentation, dilation corrosion, morphological transformation and other operations on the image through methods such as color model conversion, channel segmentation and data enhancement transformation, and uses Labeling software to mark the connecting rod surface defects, thereby obtaining a connecting rod image with the background removed and the defects marked. The model training module targets the different appearance conditions of connecting rods of different batches and models, the smaller defect sizes on the surface, and the difficulty in extracting and identifying defect features. It applies the characteristics of fast speed of one-stage target detection algorithm and high precision of two-stage target detection algorithm in target detection technology. It uses one-stage target detection algorithms such as YOLO and SSD, and two-stage target detection algorithms such as R-CNN, Fast-RCNN, and Faster-RCNN to carry out connecting rod surface defect recognition training and form a connecting rod surface defect data set (model); data enhancement methods are used to enhance the diversity and richness of the training data set, enhance the generalization ability of the model and alleviate overfitting in the training data; the data set is divided into training set, verification set, and test set in a ratio of 8:1:1 for model training, verification, and testing. The defect recognition algorithm module uses image template matching algorithm and deep learning algorithm to identify and analyze connecting rod surface defects. The PLC ladder diagram control module uses PLC and computer for ladder diagram programming to control photoelectric sensors, motors, camera external trigger photography, push rods and other mechanisms to realize automatic feeding, continuous image acquisition and recognition, and automatic grading. The human-computer interaction module mainly includes functions such as system start and stop control, product information selection, and identification of product information statistics. It is designed using Pyqt5 and can ultimately easily achieve automation, intelligence, and integrated operations.

[0012] The above description is only a preferred embodiment of the present invention, so all equivalent changes or modifications made according to the structure, characteristics and principles described in the scope of the patent application of the present invention are included in the scope of the patent application of the present invention.

Claims

1. An automatic inspection device for the surface quality of a precision forged connecting rod, comprising an automatic feeding unit, a posture adjustment unit, a shadowless lighting unit, an image acquisition unit, and a grading unit.

2. The automatic surface quality inspection equipment according to claim 1 is characterized in that The hardware system 1 mainly includes an automatic feeding unit 11 , a posture adjustment unit 12 , a shadowless lighting unit 13 , an image acquisition unit 14 , a connecting rod grading unit 15 , and a computer and control unit 16 .

3. The automatic surface quality inspection equipment according to claim 1 is characterized in that The software system 2 mainly includes an image processing module 21 , a model training module 22 , a defect recognition algorithm module 23 , a PLC ladder diagram control module 24 and a human-computer interaction module 25 .

Images