CN-122024297-A - Art work exhibition method and system based on machine vision

CN122024297ACN 122024297 ACN122024297 ACN 122024297ACN-122024297-A

Abstract

The application relates to the technical field of machine vision and discloses a method and a system for exhibition of an art work based on machine vision, wherein the method comprises the steps of firstly detecting whether a user enters a preset watching area through a human body induction sensor of an art work exhibiting area; when a user is determined to enter a preset watching area, video stream data of the watching area are acquired in real time through an image acquisition device of the showing area, real-time analysis is carried out on the video stream data, target images at least comprising the face area of the user are extracted from current video stream frames and preceding video frames, face recognition processing is carried out on the target images, recognition results comprising age characteristics and facial expression characteristics of the user are generated, finally, a dynamic showing mode corresponding to the artwork is triggered based on the recognition results, and the dynamic showing mode at least comprises one of controlling a lighting device to change the irradiation state, controlling a display device to play relevant expansion content and controlling a mechanical device to adjust the physical showing state of the artwork.

Inventors

YE ZHENHUA
YANG SHUAI
FANG XIAOYING
Wang Jiaxinhua
HUANG JIPING

Assignees

杭州遇渐文化艺术科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260123

Claims (12)

1. A machine vision-based art exhibition method, the method comprising: detecting whether a user enters a preset watching area or not through a human body induction sensor of an artwork display area; Under the condition that a user enters a preset watching area is determined, video stream data of the watching area are obtained in real time through an image acquisition device of an artwork display area; analyzing the video stream data in real time, and extracting a target image at least comprising a user face area from a current video stream frame and a preamble video frame; Performing face recognition processing on the target image to generate a recognition result, wherein the recognition result comprises age characteristics of a user and facial expression characteristics of the user; And triggering a dynamic exhibition mode corresponding to the artwork based on the identification result, wherein the dynamic exhibition mode comprises at least one of controlling a lighting device to change the irradiation state of the artwork, controlling a display device to play the expansion content related to the artwork and controlling a mechanical device to adjust the physical exhibition state of the artwork.
2. The machine vision-based artwork display method of claim 1, wherein the performing face recognition processing on the target image to generate a recognition result comprises: Selecting a preset number of continuous frame target images from the extracted target images as a concentration analysis sample set, wherein the sample set covers the face state of a user in a preset time period after entering a viewing area; Extracting expression key point coordinate data of eyebrow, eye, mouth and facial muscles for each frame of target image in the concentration analysis sample set; obtaining characteristic parameters related to the concentration expression intensity based on the expression key point coordinate data, wherein the characteristic parameters comprise one or more of eye opening and closing degree, line of sight direction, eyebrow state and facial muscle tightness; performing time sequence calibration and fusion analysis on the characteristic parameters of the multi-frame target image, and calculating the consistency duty ratio of concentrating on the relevant characteristic parameters in the multi-frame samples; Comparing the consistency duty ratio with a preset concentration threshold, if the consistency duty ratio is higher than or equal to the preset concentration threshold, judging that the facial expression features of the user are concentrated expressions, and if the consistency duty ratio is lower than the preset concentration threshold, judging that the facial expression features of the user are non-concentrated expressions.
3. The machine vision-based artwork display method of claim 1, wherein the artwork display area is further provided with an ambient brightness sensor, and the controlling the lighting device to change the illumination state of the artwork comprises: The illumination brightness is dynamically adjusted according to the environmental brightness data of the display area acquired by the environmental brightness sensor and the age characteristics of the user in the face recognition result, and/or, And dynamically adjusting an illumination mode according to facial expression characteristics of the user in the face recognition result, wherein the illumination mode comprises a warm light gradual change mode and a high light focusing mode.
4. The machine vision-based artwork display method of claim 3, wherein dynamically adjusting the illumination brightness according to the ambient brightness data of the display area collected by the ambient brightness sensor and the age characteristics of the user in the face recognition result comprises: If the face recognition result judges that the user is a child, adjusting the irradiation brightness to a preset child soft brightness interval; If the face recognition result judges that the user is an adult, comparing the environment brightness data with a preset brightness threshold, adjusting the irradiation brightness to a first brightness range when the environment brightness data is lower than a preset low threshold, adjusting the irradiation brightness to a second brightness range when the environment brightness data is higher than a preset high threshold, and adjusting the irradiation brightness to a third brightness range when the environment brightness data is between the preset low threshold and the high threshold, wherein the first brightness range is larger than the third brightness range, and the third brightness range is larger than the second brightness range.
5. The machine vision based artwork display method of claim 3, wherein the display area comprises an inner circle area and an outer circle area, the dynamically adjusting the illumination pattern according to the facial expression features of the user in the face recognition result comprises: Determining facial expression characteristics in the face recognition result as focused expressions, and controlling the lighting device to switch to a highlight focusing mode, wherein the highlight focusing mode comprises any one of highlight focusing only on an inner ring area, highlight focusing only on an outer ring area and highlight focusing alternately and circularly on the inner ring area and the outer ring area; And determining facial expression characteristics in the face recognition result to be non-attentive expressions, and controlling the lighting device to switch to a warm light gradual change mode, wherein the warm light gradual change mode comprises the step of controlling the brightness of the lighting device to circularly and gently change within a preset range of a basic value.
6. The machine vision based artwork display method of claim 1, wherein prior to the controlling the lighting device to change the state of illumination of the artwork, further comprising: acquiring a current actual heat dissipation capacity value of the lighting device, and judging whether the actual heat dissipation capacity value is lower than a preset heat dissipation threshold value; If the actual heat radiation capacity value is not lower than a preset heat radiation threshold value, directly executing the operation of controlling the lighting device to change the irradiation state; And if the actual heat radiation capacity value is lower than a preset heat radiation threshold value, controlling the rotation driving unit of the lighting device to drive the lighting substrate to rotate until the heat radiation capacity value of the lighting device reaches or is higher than the preset heat radiation threshold value, and then executing the operation of controlling the lighting device to change the irradiation state.
7. The machine vision based artwork display method of claim 6, wherein said obtaining a current actual heat dissipation capability value of said lighting device comprises: acquiring real-time temperature data of a heat dissipation assembly acquired by a temperature sensor arranged in the lighting device, and acquiring working current data of the lighting device by a current sensor; based on the temperature and the working current of the heat dissipation assembly, a current actual heat dissipation capacity value is obtained through a preset heat dissipation capacity calculation model, wherein the heat dissipation capacity calculation model meets the following expression: Heat dissipation capacity value=k (T 0 -T 1 )/I, where T 0 is the highest temperature tolerance of the heat dissipation assembly, T 1 is real-time temperature data, I is working current data, and K is a calibration coefficient related to ambient temperature data.
8. The machine vision based artwork display method of claim 6, wherein controlling the rotation driving unit to rotate the illumination substrate comprises: Acquiring an initial attitude angle of the illumination substrate, and judging whether the initial attitude angle is positioned at the end point position of the rotation range; If the initial attitude angle is positioned at a non-end point position of the rotation range, executing a rotation direction selection strategy to obtain a target rotation direction; controlling a rotary driving unit to drive the illumination substrate to rotate based on the target rotation direction; the rotation direction selection strategy includes: Acquiring a first residual angle of the lighting device, which is rotated to a first mechanical limit along a first candidate direction and a second residual angle of the lighting device, which is rotated to a second mechanical limit along a second candidate direction, from the initial attitude angle; comparing the first remaining angle with the second remaining angle; According to the comparison result, the candidate direction with the larger remaining angle is determined as the target rotation direction.
9. The machine vision based artwork display method of claim 8, wherein the rotation direction selection strategy further comprises: and under the condition that the first residual angle is the same as the second residual angle, acquiring the accumulated rotation times or the accumulated rotation angles of the motor in the clockwise and anticlockwise directions in the rotation driving unit, and selecting the direction with smaller historical accumulated rotation amount as the target rotation direction.
10. The machine vision-based artwork display method of claim 6, wherein the controlling the rotation driving unit of the lighting device to rotate the lighting substrate until the heat dissipation capacity value of the lighting device reaches or exceeds a preset heat dissipation threshold comprises: Controlling the rotation driving unit to rotate from an initial attitude angle according to a preset rotation direction and a preset stepping angle; After rotating one stepping angle each time, acquiring a heat radiation capacity value under the current attitude angle, and accumulating the length of a rotating path; If the current heat dissipation capacity value is greater than or equal to the preset heat dissipation capacity threshold value and the accumulated rotation path length is greater than the preset path length, determining the current attitude angle as a target attitude angle and stopping rotation; if the current heat dissipation capacity value is greater than or equal to the preset heat dissipation capacity threshold value and the accumulated rotation path length is less than or equal to the preset path length, recording the current attitude angle and the corresponding heat dissipation capacity value thereof, and continuing to rotate to obtain the target attitude angle.
11. The machine vision based artwork display method of claim 10, wherein the step of continuing to rotate to obtain the target attitude angle comprises: In the continuous rotation process, when the rotation path length reaches the preset path length, a posture angle with the heat radiation capacity value larger than the recorded maximum heat radiation capacity value is not found yet, and the posture angle with the maximum heat radiation capacity value is selected as a target posture angle from the recorded posture angles with the heat radiation capacity value larger than or equal to the preset heat radiation capacity threshold value; And in the continuous rotation process, when the rotation path length reaches the preset path length, finding an attitude angle with the heat radiation capacity value larger than the recorded maximum heat radiation capacity value, and taking the attitude angle corresponding to the maximum heat radiation capacity value as a target attitude angle.
12. An artwork display system, comprising: A memory for storing program codes, and A processor for invoking the program code to perform the method of any of claims 1 to 11.

Description

Art work exhibition method and system based on machine vision Technical Field The invention relates to the technical field of machine vision, in particular to an art work exhibition method and system based on machine vision. Background The traditional art exhibition modes, such as art museums, museums and the like, mostly adopt static and unified exhibition modes. Namely, an artwork (such as painting and sculpture) is displayed on a fixed illumination, fixed angle and fixed exhibition stand, and unified words or voice explanation is attached. This mode, while simple to manage, has a core drawback of lacking dynamic interaction and personalized response with the viewer. The ages, cognitive backgrounds, points of interest and emotional states of the spectators are quite different, while static exhibition modes cannot perceive and adapt to these individual differences, resulting in limited immersion, depth of understanding and interactive experience for the spectators. Therefore, the core problem faced by the current exhibition technology is how to construct a set of intelligent system capable of capturing and analyzing the biological characteristics and behavior states of audiences in real time and dynamically and accurately regulating and controlling the exhibition mode of the works of art according to the biological characteristics and behavior states. The requirement breaks through the existing unidirectional and static exhibition logic, and forms a perceivable, analyzable and responsive interactive exhibition method, thereby truly realizing bidirectional intelligent dialogue between artwork and audience, and improving the attraction, education effect and artistic appeal of exhibition. Disclosure of Invention The invention mainly aims to provide a machine vision-based art work exhibition method and system, and aims to solve the technical problems that in the prior art, a static exhibition mode cannot sense and adapt to individual differences, so that immersion feeling, understanding depth and interaction experience of audiences are limited. To achieve the above object, in a first aspect, an embodiment of the present application provides a machine vision-based art work exhibition method, including: detecting whether a user enters a preset watching area or not through a human body induction sensor of an artwork display area; Under the condition that a user enters a preset watching area is determined, video stream data of the watching area are obtained in real time through an image acquisition device of an artwork display area; analyzing the video stream data in real time, and extracting a target image at least comprising a user face area from a current video stream frame and a preamble video frame; Performing face recognition processing on the target image to generate a recognition result, wherein the recognition result comprises age characteristics of a user and facial expression characteristics of the user; And triggering a dynamic exhibition mode corresponding to the artwork based on the identification result, wherein the dynamic exhibition mode comprises at least one of controlling a lighting device to change the irradiation state of the artwork, controlling a display device to play the expansion content related to the artwork and controlling a mechanical device to adjust the physical exhibition state of the artwork. In one possible implementation manner, the performing face recognition processing on the target image to generate a recognition result includes: Selecting a preset number of continuous frame target images from the extracted target images as a concentration analysis sample set, wherein the sample set covers the face state of a user in a preset time period after entering a viewing area; Extracting expression key point coordinate data of eyebrow, eye, mouth and facial muscles for each frame of target image in the concentration analysis sample set; obtaining characteristic parameters related to the concentration expression intensity based on the expression key point coordinate data, wherein the characteristic parameters comprise one or more of eye opening and closing degree, line of sight direction, eyebrow state and facial muscle tightness; performing time sequence calibration and fusion analysis on the characteristic parameters of the multi-frame target image, and calculating the consistency duty ratio of concentrating on the relevant characteristic parameters in the multi-frame samples; Comparing the consistency duty ratio with a preset concentration threshold, if the consistency duty ratio is higher than or equal to the preset concentration threshold, judging that the facial expression features of the user are concentrated expressions, and if the consistency duty ratio is lower than the preset concentration threshold, judging that the facial expression features of the user are non-concentrated expressions. In one possible implementation manner, the artwork display area is further provided wi