Smoke and strong light interference resistant visual camera for monitoring laser welding process
By employing technologies such as multi-band image sensors and active polarization filters, the problem of unclear imaging during laser welding has been solved, enabling high-quality image acquisition and welding quality assessment even in harsh environments.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-09
- Publication Date
- 2026-04-07
AI Technical Summary
Traditional machine vision systems are subject to interference from high-intensity arc light, welding fumes, spatter particles, and infrared heat radiation during laser welding, resulting in unclear imaging and affecting the accuracy of welding quality assessment.
Employing a multi-band image sensor and active polarization filter, combined with a narrow-band interference filter and laser synchronization triggering circuit, it generates a fused image resistant to smoke interference, filters strong laser arc light and irrelevant band interference, performs image processing through an embedded GPU or FPGA chip, and integrates a thermal barrier housing to enhance stability.
It achieves stable imaging in environments with strong lasers and dense smoke, has a short dynamic response time, and high image quality, making it suitable for deployment in industrial welding systems. It also features low power consumption and high environmental adaptability.
Smart Images
Figure CN224097757U_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The utility model relates to industrial image processing technical field especially relates to a kind of anti-smoke and strong light interference vision camera for laser welding process monitoring, it is applicable to laser welding quality on-line detection and defect identification scene. BACKGROUND
[0002] In high-precision manufacturing industry, laser welding is widely used in the metal connection process of automobile, aviation, 3C electronic and other industries with its non-contact, high energy density and high automation degree advantages. However, the high-intensity arc light, welding dust, splashing particles and infrared thermal radiation generated in the laser welding process will seriously interfere with the identification of the details of the welding pool, weld formation and splashing behavior by image acquisition equipment, thereby affecting the accuracy of real-time monitoring and quality evaluation.
[0003] The traditional machine vision system is difficult to realize clear imaging in such environment due to the limited dynamic range of photosensitive elements, large imaging delay and weak anti-interference ability, and usually needs to rely on complex optical filters, gas exhaust equipment or optical isolation devices, but these schemes often have problems such as high cost, integration difficulty and complex maintenance.
[0004] Therefore, there is an urgent need for a vision camera device with high dynamic range imaging capability, strong light interference and smoke scattering capability to improve the imaging quality and reliability in the welding process. SUMMARY
[0005] To solve the above technical problems, the utility model designs an anti-smoke and strong light interference vision camera for laser welding process monitoring to solve the clear imaging problem of existing vision system in laser welding process.
[0006] The utility model adopts the following technical scheme:
[0007] An anti-smoke and strong light interference vision camera for laser welding process monitoring, the vision camera includes camera lens and camera controller, the camera lens is controlled by the camera controller, the vision camera further includes multi-band image sensor and active polarization filter, the multi-band image sensor includes visible light sensor and near-infrared sensor, the multi-band image sensor is used for synchronously collecting visible light image and near-infrared image of welding area, the visible light sensor, near-infrared sensor and active polarization filter are communicated with the camera controller. Through synchronous collection of visible light image and near-infrared image of welding area feedback to camera controller, multi-band fusion processing is carried out, and fusion image of anti-smoke interference is generated, while adapting to the spectral characteristics of different welding materials, filtering strong laser arc light and irrelevant band interference.
[0008] As preferred, the active polarization filter employs controllable liquid crystal polarizer.
[0009] As preferred, the visual camera further comprises laser synchronization trigger circuit in communication with the camera controller.
[0010] As preferred, the camera lens employs narrow-band interference filter combined with coated lens.
[0011] As preferred, the multi-band image sensor is configured with band-pass filter slot.
[0012] As preferred, the camera controller is integrated with embedded GPU or FPGA chip.
[0013] As preferred, the visual camera is externally provided with heat resistance cover.
[0014] As preferred, the camera lens is provided with high-temperature shielding glass in front.
[0015] The beneficial effects of the present application are as follows: (1) the visible light image and near-infrared image of the welding area are fed back to the camera controller for multi-band fusion processing to generate a fusion image resistant to smoke interference, and the spectrum characteristics of different welding materials are adapted, strong laser arc light and irrelevant band interference are filtered, and stable imaging can be realized under strong laser interference and dense smoke; (2) the camera lens employs narrow-band interference filter combined with coated lens to effectively filter the main wavelength of laser and high-frequency scattering; (3) the laser synchronization trigger circuit is provided to synchronize with the laser welding machine to control the shutter window and avoid strong light peaks, and the dynamic response time is short, and millisecond-level shutter control can be realized. BRIEF DESCRIPTION OF DRAWINGS
[0016] Figure 1 is a structural schematic view of the present application;
[0017] In the figure: 1, visual camera, 2, camera lens, 3, camera controller, 4, active polarization filter, 5, multi-band image sensor, 6, laser synchronization trigger circuit. DETAILED DESCRIPTION
[0018] The technical scheme of the present application will be further described in detail below by means of specific embodiments and in combination with the drawings:
[0019] Embodiment: as Figure 1As shown, a smoke and strong light interference resistant visual camera for laser welding process monitoring, the visual camera 1 comprises a camera lens 2 and a camera controller 3, the camera lens is controlled by the camera controller, the visual camera further comprises a multi-band image sensor 5 and an active polarization filter 4, the multi-band image sensor comprises a visible light sensor and a near-infrared sensor, the multi-band image sensor is used for synchronously collecting visible light images and near-infrared images of a welding area, the visible light sensor, the near-infrared sensor and the active polarization filter are in communication with the camera controller. Through synchronous collection of visible light images and near-infrared images of the welding area, feedback to the camera controller, multi-band fusion processing is carried out, a smoke interference resistant fusion image is generated, and the spectral characteristics of different welding materials are adapted, and strong laser arc light and irrelevant band interference are filtered.
[0020] The active polarization filter adopts a controllable liquid crystal polarization plate. The orientation of the liquid crystal molecules is controlled by voltage, and the optical polarization direction is adjusted in real time. Strong reflection is further suppressed.
[0021] The visual camera further comprises a laser synchronization trigger circuit 6, which is in communication with the camera controller. By synchronizing with the laser welding machine, the shutter window is controlled to avoid strong light peaks.
[0022] The camera lens adopts a narrow-band interference filter combined with a coated lens. The main wavelength of the laser and high-frequency scattering are effectively filtered.
[0023] The multi-band image sensor is configured with a band-pass filter slot. The filter with a center wavelength of 450-940nm can be quickly replaced, the spectral characteristics of different welding materials are adapted, and strong laser arc light and irrelevant band interference are filtered.
[0024] An embedded GPU or FPGA chip is integrated in the camera controller. A heat-resistant cover is arranged outside the visual camera. A high-temperature shielding glass is arranged in front of the camera lens.
[0025] Compared with the prior art, the visual camera can stably image under strong laser interference and thick smoke; the dynamic response time is short, millisecond-level shutter control can be realized; the image quality is stable, which is convenient for subsequent AI detection model to identify defects; the integration degree is high, which is suitable for embedding in an industrial welding system; it has low power consumption and high environmental adaptability, and can support long-term operation in harsh environments.
[0026] The above-mentioned embodiments are only a preferred scheme of the present application, and do not limit the present application in any form. Other variants and modifications can be made without exceeding the technical scheme recited in the claims.
Claims
1. A vision camera resistant to smoke and strong light interference for monitoring laser welding processes, the vision camera comprising a camera lens and a camera controller, the camera lens being controlled by the camera controller, characterized in that, The vision camera also includes a multi-band image sensor and an active polarization filter. The multi-band image sensor includes a visible light sensor and a near-infrared sensor. The multi-band image sensor is used to simultaneously acquire visible light images and near-infrared images of the welding area. The visible light sensor, the near-infrared sensor, and the active polarization filter are connected to the camera controller.
2. The anti-smoke and strong light interference visual camera for monitoring laser welding processes according to claim 1, characterized in that, The active polarization filter uses a controllable liquid crystal polarizer.
3. The anti-smoke and strong light interference visual camera for monitoring laser welding processes according to claim 1, characterized in that, The vision camera also includes a laser synchronization triggering circuit, which is connected to the camera controller.
4. A visual camera resistant to smoke and strong light interference for monitoring laser welding processes according to claim 1, characterized in that, The camera lens uses a narrow-band interference filter combined with a coated lens.
5. A visual camera resistant to smoke and strong light interference for monitoring laser welding processes according to claim 1, characterized in that, The multi-band image sensor is equipped with a bandpass filter slot.
6. A visual camera resistant to smoke and strong light interference for monitoring laser welding processes according to claim 1, characterized in that, The camera controller integrates an embedded GPU or FPGA chip.
7. A visual camera resistant to smoke and strong light interference for monitoring laser welding processes according to claim 1, characterized in that, The vision camera is equipped with a heat-barrier cover.
8. A visual camera resistant to smoke and strong light interference for monitoring laser welding processes according to claim 1, characterized in that, A high-temperature shielding glass is installed in front of the camera lens.