Electrician goggles with shooting function

Abstract

The electrician goggles with the shooting function comprise a goggle frame, goggle legs and mounting plates, the left mounting plate and the right mounting plate are connected with a visible light camera module and an infrared camera module in a clamped mode respectively, the visible light camera module shoots through a visible light camera, a highlight image part is restrained through an internally-arranged highlight restraining module, and the infrared camera module is connected with the visible light camera module in a clamped mode. The visible light camera module and the infrared camera module which are detachable are used for shooting, different cameras are used for shooting and displaying, the infrared camera module is used for shooting through an infrared camera, a built-in infrared image processing module is used for processing, and the infrared camera module is used for displaying through the display module, the visible light camera module and the infrared camera module which are detachable are used for shooting, and different cameras are used for shooting and displaying. According to the method, the visible light image and the infrared light image are fused, the image is processed, the highlight influence during welding is reduced, the applicability to the electrician construction environment is improved, and the image processing and target tracking recognition precision and speed are synchronously improved by fusing the visible light image and the infrared light image.

Description

Technical Field

[0001] The invention relates to the technical field of electrician goggles, in particular to an electrician goggles with a shooting function. Background Art

[0002] Welders rely on protective glasses to protect their eyes from the flash of light and the impact of heat. During the operation, the scattered particles and debris, sparks and heat flow, dazzling light and smoke can cause damage to the eyes. Large areas of pixels in the image are covered by strong original halos, which brings extreme discomfort to visual observation and even causes short-term blindness.

[0003] A Chinese patent discloses a goggles system for high-altitude work in power construction (publication number CN 110109264 A). This patented technology can free hands, take real-time photos of on-site working conditions, and transmit them to the image display platform on the ground. However, highlights and flickers during welding will affect welding accuracy, and welding operation records are incomplete. Therefore, the present invention provides an electrician goggles with a shooting function to solve the problems raised in the above background technology. Summary of the invention

[0004] The purpose of the present invention is to provide an electrician's goggles with a photographing function to solve the problem that it is inconvenient for electric workers to take photos during construction.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] The invention discloses an electrician's goggles with a shooting function, comprising a frame, temples and a mounting plate, wherein a visible light camera module and an infrared camera module are respectively clamped on the left and right mounting plates, a light-transmitting lens is mounted on the frame, the front side of the frame is rotatably connected to a display module, the display module covers the front side of the lens, buttons are arranged on the outsides of the visible light camera module and the infrared camera module, both the visible light camera module and the infrared camera module can be displayed through the display module, the visible light camera module shoots through a visible light camera, and a highlight image part is suppressed through a built-in highlight suppression module, the infrared camera module shoots through an infrared camera, and is processed through a built-in infrared image processing module, and the display module displays infrared images, visible light images and fused images.

[0007] Preferably, the front and rear cross-sections of the mounting plate are inverted L-shaped, a downward limit plate is provided on the outer side of the upper end of the mounting plate, and a horizontal clamping plate is provided on the lower end of the mounting plate. The visible light camera module and the infrared camera module are fixed on the mounting plate through the clamping plate and the limit plate.

[0008] Preferably, the rear end of the clamping plate is provided with an anti-dropping block, and the upper side of the clamping plate is provided with a plurality of protrusions for limiting movement.

[0009] Preferably, a semicircular guide groove is provided at the rear end of the temple, and an extension frame is slid in the guide groove. The side of the extension frame is pressed against the protective surface of the guide groove to improve stability after movement. A protruding toggle block is provided on the rear side of the upper end of the extension frame to adjust the position of the extension frame. An upwardly tilted elastic block is provided at the lower end of the extension frame to press under the ear to improve stability.

[0010] Preferably, the visible light camera module and the infrared camera module are connected by a detachable wire, and the infrared camera module is connected to a control box by a wire. A battery and a detachable memory card are provided in the control box, and a plug connector is provided at the end of the wire. The visible light camera module, the infrared camera module and the control box are all provided with corresponding jacks for power supply.

[0011] Preferably, a plug connector is provided in the middle of the display module, which can be plugged into the slot of the frame for installation. A protrusion is fixed on the side of the display module, and a corresponding groove is provided on the mounting plate. The position of the display module is fixed by snapping the protrusion into the groove.

[0012] The button is short pressed to turn on and off, and the image is displayed through the display module for 3-6 seconds when it is long pressed. After more than 6 seconds, the display module outputs a fused image. The fused image fuses the two images to achieve accurate RGBT tracking of the target.

[0013] Preferably, the highlight suppression module performs feature analysis on such images, extracts brightness components, reverses the original image and the brightness components, compares the degree of change of the brightness components, and returns the brightness components to the image after smoothing and filtering, thereby achieving the effect of suppressing the brightness of local strong light areas of the image. Finally, the edge texture of brightness changes in the image is restored to a certain extent through improved guided filtering, thereby significantly improving the subjective visual experience of the image.

[0014] Preferably, the guided filtering is to obtain effective attenuation of the brightness of the strong light area in the image after being processed by the brightness attenuation model, combining the algorithm operation efficiency and time, using the original image containing all image edge detail information as the guided image, and the attenuated image processed by the brightness attenuation model as the input image, and performing guided filtering on the image after brightness attenuation through parameters matching the image size. Through the edge texture information in the original image, the obvious dividing line caused by attenuation in the brightness attenuation graph is enhanced, and the original edge information near the dividing line is attached to the output image from the guided image, thereby improving the visual effect of the enhanced graph to a certain extent.

[0015] Preferably, the infrared image processing module adopts a lightweight MobileNetV architecture to strengthen infrared feature acquisition and action visual rhythm, solve the problems of weak texture features and similar behavioral features in the tracking process, and uses the K.Means clustering method to perform clustering analysis on the bounding boxes in the infrared image data set. The activation function is the ELU activation function to improve the detection effect of the operating target.

[0016] Preferably, the fusion process comprises the following steps:

[0017] Firstly, feature-level fusion is performed based on the features of infrared and visible light images to extract targets of different scales in the source image, and more complete salient target areas in the multispectral source image are detected as much as possible. The initial image, contrast image, Gabor filter processing map, and spatial and texture correlation feature maps are used as the input of the four channels of quaternion Fourier transform. The input is subjected to hyper-Fourier transform to construct a saliency map, and Gaussian kernel functions of different scales are used to generate saliency maps of different scales. The minimum information entropy theory is used to linearly generate saliency maps of the optimal scale to detect salient areas, which can detect more complete salient target areas in the source image as much as possible and avoid information loss.

[0018] Then, the proposed improved GrabCut algorithm is used to segment the image into salient and non-salient parts. This can obtain relatively complete salient areas of visible light and infrared images, and the two parts are fused using different fusion strategies to obtain the final result, thereby achieving simultaneous improvement in the accuracy and speed of image processing and target tracking and recognition.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] 1. The present invention uses a detachable visible light camera module and an infrared camera module to shoot, uses different cameras to shoot and display, and processes the image, thereby reducing the impact of high light during welding and improving the applicability to the electrical construction environment.

[0021] 2. The present invention achieves simultaneous improvement in the accuracy and speed of image processing and target tracking and recognition by fusing visible light and infrared light images. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 The figure is a schematic diagram of the structure of electrician goggles with a photographing function.

[0023] Figure 2 The present invention is a schematic diagram of the structure of the unfolded display module in an electrician's goggles with a photographing function.

[0024] Figure 3 The figure is a schematic diagram of the explosion structure of an electrician's goggles with a photographing function.

[0025] Figure 4 This is a diagram of the image fusion processing algorithm for electrician goggles with shooting function.

[0026] In the figure: 1. Spectacle frame; 2. Spectacle legs; 21. Extension frame; 22. Toggle block; 3. Mounting plate; 31. Infrared camera module; 32. Visible light camera module; 4. Display module. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] See also Figures 1 to 4 In an embodiment of the present invention, an electrician's goggles with a shooting function include a frame 1, temples 2 and a mounting plate 3. The left and right mounting plates 3 are respectively connected with a visible light camera module 32 and an infrared camera module 31. The rear end of the temple 2 is provided with a semicircular guide groove, and an extension frame 21 is slidable in the guide groove. The side of the extension frame 21 is pressed on the protective surface of the guide groove to improve the stability after movement. A protruding toggle block 22 is provided on the rear side of the upper end of the extension frame 21 for adjusting the position of the extension frame 21. An elastic block is provided at the lower end of the extension frame 21, which is tilted upward and pressed under the ear to improve stability. A light-transmitting lens is installed on the frame 1, and the front side of the frame 1 is rotatably connected to the display module 4. The display module 4 covers the front side of the lens, and a plug connector is provided in the middle of the display module 4, which can be plugged into the slot of the frame 1 for installation. The front and rear cross-sections of the mounting plate 3 are inverted L-shaped, and a downward limit plate is provided on the outer side of the upper end of the mounting plate 3, and a horizontal clamping plate is provided at the lower end of the mounting plate 3. The visible light camera module 32 and the infrared camera module 31 are fixed to the mounting plate 3 through the clamping plate and the limit plate. The rear end of the clamping plate is provided with an anti-slip block, and a plurality of protrusions are provided on the upper side of the clamping plate for limiting movement. The outer sides of the visible light camera module 32 and the infrared camera module 31 are provided with buttons, and both the visible light camera module 32 and the infrared camera module 31 can be displayed through the display module 4.

[0029] The visible light camera module 32 and the infrared camera module 31 are connected by a detachable wire, and the infrared camera module 31 is connected to a control box by a wire. A battery and a detachable memory card are provided in the control box, and a plug connector is provided at the end of the wire. The visible light camera module 32, the infrared camera module 31 and the control box are all provided with corresponding jacks for power supply.

[0030] The visible light camera module 32 shoots through a visible light camera, suppresses the highlight image part through a built-in highlight suppression module, and displays it through the display module 4. The infrared camera module 31 shoots through an infrared camera, processes it through a built-in infrared image processing module, and displays it through the display module 4.

[0031] A convex block is fixed on the side of the display module 4, and a corresponding groove is provided on the mounting plate 3. The position of the display module 4 is fixed by snapping the convex block into the groove. The button is short pressed to turn on and off the machine, and the image is displayed through the display module 4 for 3-6 seconds when long pressed. After more than 6 seconds, the display module 4 outputs a fused image. The fused image fuses the two images to achieve accurate RGBT tracking of the target.

[0032] The highlight suppression module performs feature analysis on such images, extracts brightness components, reverses the original image and brightness components, compares the degree of change of the brightness components, and returns the brightness components to the image after smoothing and filtering, thereby achieving the effect of suppressing the brightness of the local strong light area of ​​the image. Finally, the edge texture of the brightness change in the image is restored to a certain extent through the improved guided filtering, thereby significantly improving the subjective visual experience of the image.

[0033] The guided filtering is to effectively attenuate the brightness of the strong light area in the image after being processed by the brightness attenuation model, combine the algorithm operation efficiency and time, use the original image containing all image edge detail information as the guided image, and use the attenuated image processed by the brightness attenuation model as the input image. After the parameters matching the image size, the image after brightness attenuation is subjected to guided filtering processing. Through the edge texture information in the original image, the obvious dividing line caused by attenuation in the brightness attenuation graph is enhanced, and the original edge information near the dividing line is attached to the output image from the guided image, thereby improving the visual effect of the enhanced graph to a certain extent.

[0034] The infrared image processing module adopts a lightweight MobileNetV4 architecture to enhance infrared feature acquisition and action visual rhythm, solve the problems of weak texture features and similar behavioral features in the tracking process, and uses the K.Means clustering method to perform clustering analysis on the bounding boxes in the infrared image data set. The activation function is the ELU activation function to improve the detection effect of the operating target.

[0035] The fusion process comprises the following steps:

[0036] Firstly, feature-level fusion is performed based on the features of infrared and visible light images to extract targets of different scales in the source image, and more complete salient target areas in the multispectral source image are detected as much as possible. The initial image, contrast image, Gabor filter processing map, and spatial and texture correlation feature maps are used as the input of the four channels of quaternion Fourier transform. The input is subjected to hyper-Fourier transform to construct a saliency map, and Gaussian kernel functions of different scales are used to generate saliency maps of different scales. The minimum information entropy theory is used to linearly generate saliency maps of the optimal scale to detect salient areas, which can detect more complete salient target areas in the source image as much as possible and avoid information loss.

[0037] Then, the proposed improved GrabCut algorithm is used to segment the image into salient and non-salient parts. This can obtain relatively complete salient areas of visible light and infrared images, and the two parts are fused using different fusion strategies to obtain the final result, thereby achieving simultaneous improvement in the accuracy and speed of image processing and target tracking and recognition.

[0038] When encountering drastic changes in the detection image, the multispectral image is constrained by spatiotemporal regularization. By obtaining effective regularization weights, the regularization will change adaptively when encountering different scenes, ensuring that the target can still be stably tracked when the target color and background appearance change. Therefore, it is robust to most challenging scenes (such as illumination changes and occlusions). Visible light and thermal infrared modal information are fused through an adaptive weighting scheme, and the model is updated using a high-confidence relative peak discrimination update method. This avoids model drift when the image target shakes violently and reduces the impact of erroneous model updates on the tracking process.

[0039] The working principle of the present invention is:

[0040] When in use, the infrared camera module 31 and the visible light camera module 32 are connected by the mounting plate 3. In daily use, the display module 4 is flipped upwards and the protrusion is inserted into the groove. When welding is required, the display module 4 is put down, and the image is captured by the visible light camera module 32 and the infrared camera module 31, and the image is displayed on the display module 4. Different pictures can be selected as needed.

[0041] Short press the button to turn it on, and long press the button to transmit the image on that side to the display module 4. After long pressing for more than 6 seconds, the fused image will be transmitted, which is convenient for operation.

[0042] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An electrician's goggles with a photographing function, comprising a frame (1), temples (2) and a mounting plate (3), characterized in that: The left and right mounting plates (3) are respectively connected with a visible light camera module (32) and an infrared camera module (31); a light-transmitting lens is mounted on the lens frame (1); the front side of the lens frame (1) is rotatably connected to a display module (4); the display module (4) covers the front side of the lens; buttons are arranged on the outsides of the visible light camera module (32) and the infrared camera module (31); both the visible light camera module (32) and the infrared camera module (31) can be displayed through the display module (4); the visible light camera module (32) takes pictures through a visible light camera, and suppresses the highlight image part through a built-in highlight suppression module; the infrared camera module (31) takes pictures through an infrared camera, and processes the pictures through a built-in infrared image processing module; the display module (4) displays the infrared image, the visible light image and the fused image.

2. The electrician goggles with a photographing function according to claim 1, characterized in that: The front and rear cross-sections of the mounting plate (3) are inverted L-shaped, a downward limiting plate is provided on the outer side of the upper end of the mounting plate (3), and a horizontal clamping plate is provided on the lower end of the mounting plate (3), and the visible light camera module (32) and the infrared camera module (31) are fixed to the mounting plate (3) via the clamping plate and the limiting plate.

3. The electrician goggles with a photographing function according to claim 2, characterized in that: An anti-dropping block is arranged at the rear end of the clamping plate, and a plurality of protrusions are arranged on the upper side of the clamping plate.

4. The electrician goggles with a photographing function according to claim 1, characterized in that: A semicircular guide groove is provided at the rear end of the temple (2), and an extension frame (21) is slidable in the guide groove. The side surface of the extension frame (21) is pressed against the protective surface of the guide groove to improve stability after movement. A protruding toggle block (22) is provided at the rear side of the upper end of the extension frame (21), and an upwardly tilted elastic block is provided at the lower end of the extension frame (21).

5. The electrician goggles with a photographing function according to claim 1, characterized in that: The visible light camera module (32) and the infrared camera module (31) are connected via a detachable wire, and the infrared camera module (31) is connected to a control box via a wire. A storage battery and a detachable memory card are arranged in the control box. A plug connector is arranged at the end of the wire, and corresponding sockets are arranged on the visible light camera module (32), the infrared camera module (31) and the control box.

6. The electrician goggles with a photographing function according to claim 1, characterized in that: A plug connector is provided in the middle of the display module (4), and the plug connector can be plugged into the slot of the mirror frame (1). A convex block is fixed on the side of the display module (4), and a corresponding groove is provided on the mounting plate (3).

7. The electrician goggles with a photographing function according to claim 1, characterized in that: The highlight suppression module performs feature analysis on such images, extracts brightness components, reverses the original image and brightness components, compares the degree of change of brightness components, returns the brightness components to the image after smoothing and filtering, and finally restores the edge texture of brightness changes in the image through improved guided filtering.

8. The electrician goggles with a photographing function according to claim 7, characterized in that: The guided filtering is to effectively attenuate the brightness of the strong light area in the image after being processed by the brightness attenuation model, combine the algorithm operation efficiency and time, use the original image containing all image edge detail information as the guided image, use the attenuated image processed by the brightness attenuation model as the input image, and perform guided filtering on the image after brightness attenuation through parameters matching the image size. Through the edge texture information in the original image, the obvious boundary line caused by attenuation in the brightness attenuation graph is strengthened, and the original edge information near the boundary line is attached to the output image from the guided image.

9. The electrician goggles with a photographing function according to claim 1, characterized in that: The infrared image processing module adopts a lightweight MobileNetV4 architecture to enhance infrared feature acquisition and action visual rhythm, solve the problems of weak texture features and similar behavioral features in the tracking process, and uses the K.Means clustering method to perform clustering analysis on the bounding boxes in the infrared image dataset, with the activation function being the ELU activation function.

10. The electrician goggles with a photographing function according to claim 1, characterized in that: The fusion process comprises the following steps: Firstly, feature-level fusion is performed based on the features of infrared and visible light images to extract targets of different scales in the source image. The initial image, contrast image, Gabor filter processing map, and spatial and texture correlation feature maps are used as the input of the four channels of quaternion Fourier transform. The input is subjected to hyper-Fourier transform to construct a saliency map. Gaussian kernel functions of different scales are used to generate saliency maps of different scales. The minimum information entropy theory is used to linearly generate the saliency map of the optimal scale to detect salient areas. Then, the proposed improved GrabCut algorithm is used to segment the image into a salient part and a non-salient part, and the two parts are fused using different fusion strategies to obtain the final result, thereby achieving simultaneous improvement in the accuracy and speed of image processing and target tracking recognition. When encountering drastic changes in the detection image, the multispectral image is subjected to spatiotemporal regularization constraints. By obtaining effective regularization weights, the regularization will adaptively change when encountering different scenes, ensuring that the target can still be stably tracked when the target color and background appearance change. The visible light and thermal infrared modal information are fused through an adaptive weighting scheme, and the model is updated using a high-confidence relative peak discrimination update method.

Images