Fish tank dynamic interaction display device based on position detection and interaction implementation method

Abstract

The application discloses a fish tank dynamic interactive display device and an interactive implementation method based on position detection, belongs to the field of intelligent aquarium equipment and digital interaction technology, and realizes real-time collection of fish body images, touch and voice interaction of a control screen, and dynamic update and display of content of a background display screen according to the position of the fish body. A control processing hardware module is integrated with a microprocessor, a real-time digital graphic processing chip and an artificial intelligence processing unit to realize image correction, multi-fish body detection and tracking, coordinate mapping and dynamic background rendering. The method process comprises the steps of image collection, geometric correction, position detection, coordinate mapping, dynamic rendering and cross-device linkage, supports gesture interaction, AI-generated background and multi-fish tank linkage functions. Through the combination of real-time position detection and dynamic display technology, an immersive interactive experience of fish movement scenes is constructed, the problem of insufficient interaction of traditional fish tanks is solved, and the method is suitable for home, commercial display and other scenes.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of intelligent aquarium equipment and digital interaction technology, and specifically relates to a fish tank dynamic interaction display device based on position detection and an interaction implementation method. BACKGROUND

[0002] In the field of aquarium equipment technology, traditional fish tanks have relatively single functions and are mainly used for fish feeding and basic ornamental purposes, and it is difficult to realize dynamic interaction with fish. With the development of digital technology, intelligent aquarium equipment has gradually become the industry development trend, but the existing technology still has significant deficiencies in real-time interaction and user experience.

[0003] Chinese patent CN110536086A discloses an artificial intelligence liquid crystal television fish tank, which includes a liquid crystal television and a control box, and realizes basic display functions through function modules in the control box. However, it cannot dynamically adjust the display content according to the real-time position information of the fish, and the interaction is weak, and it still belongs to the "static display + preset program" mode in essence.

[0004] The multimedia 3D simulation fish tank system disclosed in Chinese patent CN109840829A combines the client with the cloud server, but focuses on 3D effect presentation and lacks dynamic interaction mechanism with real fish, and fails to establish real-time mapping relationship between fish movement trajectory and virtual scene.

[0005] The internet social interaction electronic fish tank system described in Chinese patent CN104589888B realizes networking through a WiFi module, but its technical focus is on social interaction functions, and it lacks technical implementation in real-time visual interaction (such as fish position detection and dynamic display linkage), and cannot meet the user demand for "fish-screen real-time interaction".

[0006] The precise touch interaction method and system based on MR fish tank proposed in Chinese patent CN111897423B solves the problem of touch precision in the MR scene, but only focuses on touch offset compensation in human-computer interaction, and does not involve real-time interaction between fish and display content, and the interaction dimension is limited to the "human-device" level, lacking dynamic correlation between "device-biology".

[0007] The virtual fish tank and its implementation method disclosed in Chinese patent CN107329633A are based on laser projection and computer image generation technology, and focus on virtual effect presentation, but lack interaction mechanism with real fish, and essentially belong to a pure virtual display scheme of "no real fish tank", which cannot be applied to dynamic interaction demand in real aquarium scenes.

[0008] The core defects of the existing technology are concentrated in:

[0009] 1. Visual interaction tomography: The display system lacks real-time linkage mechanism with fish body dynamics, and cannot generate corresponding dynamic effects in real time according to the position changes of the fish, resulting in a fragmented experience of "fish moving and scene not moving". For example, when the fish swims, the background picture cannot synchronously present the linkage effects such as water grass swinging and light and shadow changing, and the viewing experience stays in the fragmented state of "static fish tank + fixed background".

[0010] 2. Weak image recognition anti-interference ability: In the fish tank environment, the background light source or the display screen reflection is easy to produce image noise points on the glass surface. The traditional camera uses a general processor to analyze the image (the processing speed is only 15-20 frames per second), which leads to the lag of fish position detection (the end-to-end delay can reach more than 200 ms), and the dynamic background update cannot match the real-time movement of the fish body.

[0011] 3. Single interaction mode: It relies on physical keys or simple touch, lacks natural interaction modes such as voice commands and gesture recognition, and the background content is limited to preset templates, which cannot generate personalized scenes according to user preferences (such as voice command "generate a submarine volcano scene"), and the interaction depth is insufficient.

[0012] 4. Lack of multi-device cooperation: In commercial display or home cluster scenarios, multiple fish tanks cannot synchronize fish school position data, and each device background display is independent, making it difficult to create an immersive overall experience.

[0013] In the prior art, the development of fish tank visual interaction function is still in the initial stage. On the one hand, the display system lacks real-time linkage mechanism with fish body dynamics, and cannot generate and display corresponding dynamic effects in real time according to the position changes of the fish; on the other hand, the human-computer interaction mode is single, and relies on preset programs or simple operations, making it difficult to realize the deep interaction between users and fish tank display content. In addition, the existing technology has obvious deficiencies in image recognition anti-interference ability, system response speed and interaction experience. In addition, the existing technology lacks support for user gesture interaction, personalized background generation and multi-device cooperative display. For example, it is difficult to dynamically adjust the background effect according to the user's gesture, it is difficult to generate customized scenes based on natural language instructions, and the fish school movement cannot be linked across devices between multiple fish tanks, limiting the expandability of the interaction scene.

[0014] Therefore, how to organically combine image recognition, dynamic display and multi-element interaction technology, and systematically solve the technical bottlenecks of traditional fish tank interaction deficiency and scene solidification, to provide a dynamic interaction solution for intelligent aquarium equipment, significantly improve user experience and product competitiveness, has become a technical problem to be solved in the field. SUMMARY

[0015] In view of this, the purpose of the present application is to overcome the defects of insufficient real-time interaction, image recognition susceptible to interference and single interaction mode of the prior art intelligent fish tank, and to provide a fish tank dynamic interaction display device and an interaction implementation method based on position detection. Through the real-time capture of fish position information by the miniature camera, the background reflection is shielded by the polarized filter to improve the image acquisition clarity, the millisecond-level response of fish detection and tracking is realized with the help of the special digital graphic processing chip and the neural network processor, and the multi-element interaction modules such as voice and gesture and the cloud AI background generation function are integrated, the real-time linkage mechanism of fish movement trajectory and dynamic background is established, thereby solving the split problem of the traditional fish tank that the fish moves and the scene does not move, realizing the closed loop of "position detection-intelligent processing-dynamic display-natural interaction", and providing the user with high immersion, high interactivity and rich scene intelligent aquarium experience.

[0016] In order to achieve the above purpose, the first aspect of the present application provides a fish tank dynamic interaction display device based on position detection, comprising:

[0017] A fish tank body;

[0018] A miniature camera for real-time acquisition of fish body images in the fish tank and output of image signals, a polarized filter is arranged in front of the lens of the miniature camera for shielding the light interference of the background display screen;

[0019] A control screen, which is a touch screen hardware unit integrated with a microphone, for displaying system status and receiving user instructions;

[0020] A background display screen installed at the rear of the fish tank body for playing dynamic backgrounds and displaying dynamic effects according to the position of the fish;

[0021] A control processing hardware module including a microprocessor, a real-time digital graphic processing chip, an artificial intelligence processing unit and a storage unit, the microprocessor is connected with the miniature camera, the control screen and the background display screen, and is used for:

[0022] Processing the image signal through the real-time digital graphic processing chip to obtain the position information of the fish;

[0023] Controlling the background display content according to the user instructions;

[0024] Converting the fish position information into an image control signal containing dynamic effect parameters and outputting the image control signal to the background display screen, wherein the real-time digital graphic processing chip integrates a geometric transformation unit and a mask generation unit for image correction and fish tank area cutting respectively, the artificial intelligence processing unit is used for executing multi-fish body detection, tracking and coordinate mapping, the control processing hardware module further includes a coordinate mapping unit and a display space transformation unit for mapping the fish body coordinates to the display coordinate system, and integrates a physical engine for calculating the dynamic background effect parameters based on the fish body position.

[0025] Further, the gesture interaction module is further included, the gesture interaction module includes a gesture recognition camera and a deep neural network processor, the gesture recognition camera is used for collecting gesture images, and the deep neural network processor is used for identifying gestures and triggering dynamic effects.

[0026] Further, the control processing hardware module further includes a multi-device networking interface, supports synchronizing fish body position data of multiple fish tanks through a local area network, and realizes cross-device background linkage.

[0027] The second aspect of the application provides a fish tank dynamic interaction implementation method based on position detection, which adopts the device described above, and the signal processing procedure of fish body position detection and dynamic background linkage includes the following steps:

[0028] S1, a miniature camera with a polarized filter is used to collect fish tank video streams, and image data is transmitted to a control processing hardware module;

[0029] S2, a real-time digital image processing chip is used to perform geometric correction, size standardization, elevation angle distortion correction and fish tank region cutting on the image signal, specifically including:

[0030] S21, a scaling factor is dynamically calculated based on image width and height parameters, and the image is compressed to a preset size;

[0031] S22, an image rotation transformation, perspective correction and scaling operation are performed by a geometric transformation unit;

[0032] S23, perspective distortion correction is performed on the elevation angle shooting picture based on the camera intrinsic parameter matrix;

[0033] S24, a rectangular cutting mask is generated by a mask generation unit to remove redundant background;

[0034] S3, a model is called by an artificial intelligence processing unit to perform multi-fish body detection, tracking and coordinate mapping to generate structured data containing fish body position coordinates and velocity vectors, specifically including:

[0035] S31, a deep separable convolution and a cross-stage residual structure are used to construct a backbone network to output multi-scale feature maps;

[0036] S32, candidate boxes are generated according to anchor box parameters, and a dynamic threshold strategy is combined with a weighted NMS algorithm to screen target boxes;

[0037] S33, the IoU threshold value is dynamically adjusted by a target density matrix to suppress redundant detection boxes;

[0038] S34, a motion-appearance dual feature correlation tracker is used to realize multi-target ID management;

[0039] S4, converting the detection coordinates into a fish position set in the background display coordinate system through a coordinate mapping unit and a display space conversion unit and a velocity vector group, specifically including:

[0040] S41, the appearance feature adopts ResNet-18 to output a 128-dimensional vector, and the motion prediction uses Kalman filtering;

[0041] S42, the detection coordinates are restored to the original resolution through batch matrix operation compensation scaling and cropping operation;

[0042] S43, mapping the fish position set in the original coordinate system to the display coordinate system;

[0043] S5, dynamic rendering: according to the structured data and user instructions, the physical engine calculates the dynamic effect parameters, generates dynamic image control signals and outputs to the background display screen, specifically including:

[0044] S6, after receiving the control signal, the background display screen optimizes the pixel response speed through the display driving chip, displays the dynamic effect to the corresponding position, and ensures that the fish movement and virtual background effect are real-time linked and visually aligned accurately.

[0045] Further, step S5 further includes:

[0046] S51, reading the current background mode and fish position / velocity data;

[0047] S52, calling a GPU-accelerated fluid dynamics model to simulate multi-ripple superposition effect, and calculating a diffusion parameter matrix based on the vertical velocity component of the fish;

[0048] S53, calculating the emission parameters of the firework particle group based on the fish velocity vector, and using spatial hashing to optimize the overlap area calculation;

[0049] S54, calculating the calculation accuracy based on the importance score of the fish, and generating an end-to-end delayed dynamic effect;

[0050] Further, step S3 further includes:

[0051] S35, performing feature extraction through a deep separable convolution backbone network;

[0052] S36, fusing multi-scale features through a feature pyramid network to improve small target detection capability;

[0053] S37, dynamic threshold screening: adaptively adjusting the confidence threshold according to the target density;

[0054] S38, trajectory tracking: based on the Kalman filtering algorithm, the fish in the continuous frame is associated and tracked.

[0055] Further, step S5 also includes:

[0056] S55, ripple effect generation: calculate multiple ripple superposition interference parameters based on the vertical velocity component of the fish body;

[0057] S56, particle system control: generate color and density differentiated firework particle group emission parameters according to the fish body velocity vector;

[0058] S57, physical engine simulation: real-time calculation of physical parameters of dynamic effects by GPU accelerated physical engine.

[0059] Further, it also includes a multi-aquarium linkage process implementation method, the specific steps are:

[0060] S61, synchronize the fish population coordinate set and device ID of each fish tank in real time through the multi-device networking interface;

[0061] S62, when the fish body in the first fish tank moves to the preset area, trigger the second fish tank background display screen to generate a cooperative dynamic effect;

[0062] S63, build a cross-device virtual ecological scene, and combine the fish population movement trajectories of multiple fish tanks into a migration animation segment display;

[0063] S64, global coordinate mapping: uniformly map the fish body position data of multiple fish tanks to a shared coordinate system;

[0064] S65, linkage effect rendering: generate dynamic background effects based on global coordinates.

[0065] Further, it also includes an AI generated background interface implementation method, the specific steps are:

[0066] S71, receive user voice instructions through the control screen, and convert them into text through a voice recognition chip;

[0067] S72, transmit the text to the cloud StableDiffusion model through the wireless communication hardware unit;

[0068] S73, download the generated 1080p@60fps dynamic video stream and superimpose it with the local fish body position data for output;

[0069] S74, identify the scene type based on the current fish body distribution and behavior pattern;

[0070] S75, send a background generation request to the cloud server through the wireless communication hardware unit;

[0071] S76, receive and cache the dynamic background materials generated by the cloud;

[0072] S77, dynamically switching the background material adapted according to the fish body position.

[0073] Further, the implementation method of gesture interaction is also included, and the specific steps are as follows:

[0074] S81, capturing a user gesture image through a gesture recognition camera;

[0075] S82, a deep neural network processor runs a MediaPipe algorithm to output 21 hand joint coordinates;

[0076] S83, mapping the gesture position (hx, hy) to a dynamic effect trigger point, and the gesture speed is positively correlated with the effect intensity;

[0077] S84, gesture feature extraction: extracting gesture features through a deep neural network processor;

[0078] S85, action classification: identifying waving, clicking and sliding actions based on a pre-trained gesture classification model;

[0079] S86, coordinate mapping: mapping gesture coordinates to a background display coordinate system;

[0080] S87, effect triggering: generating and displaying corresponding dynamic effects according to the recognized gesture action and coordinate position.

[0081] The above technical scheme is adopted in the present application, and the following beneficial effects are achieved:

[0082] 1. Real-time dynamic linkage, enhancing the interest of observation

[0083] The fish body position is captured in real time through a miniature camera, and a control processing hardware module and a background display screen are combined to realize real-time synchronous linkage of fish body movement and dynamic background (such as virtual water grass swinging and fish school interaction effect), break the fragmentation of traditional fish tank “static background + fish body independent movement”, and significantly improve the real-time and interest of the observation process.

[0084] 2. Hardware anti-interference design, improving detection accuracy

[0085] The polarization filter in front of the lens of the miniature camera shields the background display screen reflection through the physical optical filtering principle, effectively avoids image noise interference, improves the clarity of fish body image acquisition, significantly improves the position detection accuracy, and guarantees the stability and accuracy of the interaction effect.

[0086] 3. Multiple interaction modes, building a three-dimensional interaction system

[0087] The integrated microphone capacitive touch control screen supports multiple interaction methods such as touch operation and voice instruction, and users can switch background animations and virtual scenes in real time to build a three-dimensional interaction system of "user-fish-display system", break through the traditional single ornamental mode of fish tank, and improve the operation convenience and user participation.

[0088] 4. Dedicated hardware acceleration ensures system performance

[0089] The control processing hardware module uses a dedicated image processing DSP chip and high-speed hardware interfaces such as USB3.0 and HDMI to realize real-time transmission and processing of image signals, improve system response speed and anti-interference ability, and provide hardware-level protection for long-term stable operation of the device.

[0090] 5. AI empowerment for personalized interaction

[0091] The integrated AI processing unit supports simultaneous identification and tracking of multiple fish bodies through deep learning algorithms, and adjusts the background display content according to the species, quantity, and behavior patterns of the fish to provide differentiated and personalized interactive experiences.

[0092] 6. Extensive functionality for improved scene adaptability

[0093] Gesture interaction function: Trigger virtual effects (such as generating ripples by waving hands) directly through hand movements to build a more natural human-computer interaction interface and enhance user immersion;

[0094] AI-generated background function: Generate personalized dynamic scenes (such as "underwater volcano") based on voice instructions to break through the limitations of traditional preset backgrounds and meet users' diverse needs;

[0095] Multi-fish tank linkage system: Supports synchronization of fish population position data and background collaborative rendering across devices, suitable for multi-scene applications such as family cluster display and commercial scene narration, and creates a cross-device virtual ecological linkage experience.

[0096] 7. Comprehensive technological innovation to solve industry pain points

[0097] Systematically integrating image recognition, dynamic display, multi-element interaction, and multi-device collaboration technology, the system effectively solves core problems such as insufficient interaction, fixed display content, and weak anti-interference ability of traditional fish tanks, providing a high-intelligent and highly expandable solution for intelligent aquarium devices, with significant technological progress and market application value.

[0098] The application solves the core problems of the traditional fish tank, such as insufficient interaction and linkage lag, through the whole process innovation of anti-interference collection, high-precision detection, real-time rendering, cross-device cooperation and signal processing, realizes the immersive experience of 'fish moving scene following', and opens up the multi-application scene of intelligent aquarium equipment through the technical modular design, and has significant technical progress and industrial application value. BRIEF DESCRIPTION OF DRAWINGS

[0099] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description only represent some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0100] Figure 1 is the structure diagram of the fish tank dynamic interactive display control device of the application Figure 1 .

[0101] Figure 2 is the structure diagram of the fish tank dynamic interactive display control device of the application Figure 1 .

[0102] Figure 3 is the system structure diagram of the fish tank dynamic interactive display control device of the application.

[0103] Figure 4 is the flow chart of the fish tank dynamic interactive implementation method of the application.

[0104] In the figure: 1, fish tank main body, 2, miniature camera, 3, control screen, 4, background display screen, 5, control processing hardware module, 6, polarized filter, 7, microprocessor, 8, real-time digital graphics processing chip, 9, storage unit, 10, wireless communication hardware unit, 11, artificial intelligence processing unit, 12, multi-device networking interface, 13, geometric transformation unit, 14, mask generation unit, 15, coordinate mapping unit, 16, display space transformation unit, 17, physical engine, 18, gesture interaction module, 19, gesture recognition camera, 20, deep neural network processor. DETAILED DESCRIPTION

[0105] Hereinafter, exemplary embodiments will be described in detail with reference to the accompanying drawings. In the following description, the same numbers refer to the same elements throughout the drawings. The embodiments described in the following exemplary embodiments do not represent all the embodiments consistent with the application.

[0106] Embodiment one

[0107] Please refer toFigure 1 、 Figure 2 and Figure 3 As shown in the figure, the embodiment provides a fish tank dynamic interaction display device based on position detection, comprising:

[0108] a fish tank main body 1;

[0109] a miniature camera 2 for real-time collection of fish body images in the fish tank and output of image signals, wherein a polarized filter 6 is arranged in front of the lens of the miniature camera 2 for shielding of light interference of a background display screen 4;

[0110] a control screen 3 which is a touch screen hardware unit integrated with a microphone, for display of system status and reception of user instructions;

[0111] the background display screen 4 installed at the back of the fish tank main body 1, for playing of dynamic backgrounds and display of dynamic effects according to the position of the fish;

[0112] a control processing hardware module 5 including a microprocessor 7, a real-time digital graphics processing chip 8, an artificial intelligence processing unit 11 and a storage unit 9, wherein the microprocessor 7 is connected to the miniature camera 2, the control screen 3 and the background display screen 4, for:

[0113] processing of the image signals by the real-time digital graphics processing chip 8 to obtain the position information of the fish;

[0114] control of the background display content according to the user instructions;

[0115] conversion of the fish position information into image control signals containing dynamic effect parameters and output to the background display screen 4, wherein the real-time digital graphics processing chip 8 is integrated with a geometric transformation unit 13 and a mask generation unit 14 for image correction and fish tank area cropping respectively, the artificial intelligence processing unit 11 is used for execution of multi-fish body detection, tracking and coordinate mapping, the control processing hardware module 5 further includes a coordinate mapping unit 15 and a display space transformation unit 16 for mapping of the fish body coordinates to a display coordinate system, and is integrated with a physics engine 17 for calculation of dynamic background effect parameters based on the fish body position.

[0116] As an implementation mode, the embodiment further includes a gesture interaction module 18, wherein the gesture interaction module 18 includes a gesture recognition camera 19 and a deep neural network processor 20, the gesture recognition camera 19 is used for collection of gesture images, and the deep neural network processor 20 is used for recognition of gestures and triggering of dynamic effects.

[0117] As an implementation mode, the control processing hardware module 5 in the embodiment further includes a multi-device networking interface 12 for supporting synchronization of fish body position data of multiple fish tanks through a local area network and realization of cross-device background linkage.

[0118] Example 2

[0119] See also Figure 1 、 Figure 2 and Figure 3 As shown, this embodiment provides a fish tank dynamic interactive display device based on position detection and an interactive implementation method, including:

[0120] The fish tank body 1 is a transparent glass container made of 8mm thick high-transmittance tempered glass, used for raising ornamental fish;

[0121] Miniature camera 2, mounted on the front of the base of the fish tank body 1, contains a 1 / 2.3-inch CMOS high-resolution image sensor that supports 1080p@60fps video capture. It is connected to the fish tank base via a connecting rod with a built-in data cable to capture fish position information in real time and output image signals;

[0122] Polarization filter 6, installed in front of the lens of micro camera 2, is a circular polarization optical lens with a diameter of 25 mm. It shields the background display screen 4 through the principle of physical optical filtering, thereby improving the anti-interference ability of fish recognition;

[0123] The control screen 3 is installed in front of the micro camera 2 and is a 7-inch capacitive touch screen hardware unit with a resolution of 1024×600 pixels. It has a built-in MEMS microphone and voice recognition chip, and is used to display system status and input user commands through touch or voice;

[0124] The background display 4 is installed behind the fish tank body 1 and is a 1080p resolution (1920×1080p pixels) LCD display hardware module with a refresh rate of 60Hz and an integrated HDMI2.0 video receiving interface for playing dynamic backgrounds and displaying corresponding dynamic effects according to the position of the fish;

[0125] The control processing hardware module 5 is installed behind the background display screen 4, such as Figure 3 As shown, it includes a microprocessor 7, a real-time digital graphics processing chip 8, a storage unit 9, a wireless communication hardware unit 10 and an artificial intelligence processing unit 11. The microprocessor 7 is connected to the micro camera 2, the control screen 3 and the background display screen 4 through a data bus, and is used to: receive the image signal output by the micro camera 2, and obtain the position information of the fish through the real-time digital graphics processing chip 8; control the display content of the background display screen 4 through the hardware interface circuit according to the user instructions input from the control screen 3; convert the extracted fish position information into a dynamic image control signal, and output it to the background display screen 4 through the transmission interface.

[0126] In this embodiment, the microprocessor adopts the RuiCore RK3588 chip, whose CPU is an 8-core heterogeneous architecture, including 4 cores of ARM Cortex-A76 (with a maximum frequency of 2.4 GHz) and 4 cores of Cortex-A55 (with a maximum frequency of 1.8 GHz), and integrating an ARM Mali-G610MP4 GPU graphics processing unit, supporting 8K video decoding and rendering. In addition, the chip is built-in with a 6TOPS computing power neural network processor (NPU), which can realize intelligent tasks such as multi-target recognition and deep learning inference. In this embodiment, the real-time digital graphics processing chip 8 is a special image processing DSP chip, which integrates a convolutional neural network acceleration unit, supports real-time target detection algorithms, and can process 1080p resolution image data at a speed of 60 frames per second, realizing millisecond-level fish body position detection and tracking.

[0127] In this embodiment, the storage unit 9 includes 8GB DDR4 RAM and 128GB eMMC flash memory, used for storing system software, dynamic background material library and user configuration data.

[0128] In this embodiment, the wireless communication hardware unit 10 is a dual-frequency Wi-Fi 6 chip, supporting 2.4 GHz and 5 GHz frequency bands, connected with the microprocessor 7 through a UART serial interface, realizing device networking and remote control functions.

[0129] In this embodiment, the artificial intelligence processing unit 11 is a neural network processor (NPU) with a computing power of 6TOPS (60 trillion operations per second), used for executing deep learning algorithms, realizing multi-fish body simultaneous recognition and tracking, and adaptively adjusting the background display content according to the fish species, quantity and behavior mode.

[0130] In this embodiment, the micro camera 2 and the control processing hardware module 5 are connected through a USB 3.0 data interface, realizing hardware real-time transmission of fish body image signals.

[0131] In this embodiment, the control screen 3 is a capacitive touch hardware panel, supporting 10-point touch, built-in MEMS microphone and voice recognition chip, connected with the microprocessor 7 through an I2S audio interface, realizing hardware signal conversion of touch instructions and voice instructions.

[0132] In this embodiment, the dynamic background of the background display screen 4 includes artistic pictures, animation effects or virtual scenes pre-stored in the storage unit 9, which are selected and loaded to the background display screen 4 through the hardware buttons or touch interface of the control screen 3.

[0133] Embodiment three

[0134] Please refer to Figure 1 , Figure 2 ,Figure 3 and Figure 4 As shown in the figure, the embodiment provides a fish tank dynamic interaction implementation method based on position detection, which adopts the above device, and a fish position detection and dynamic background linkage signal processing flow includes the following steps:

[0135] S1, a micro camera 2 with a polarized filter 6 is used to collect fish tank video streams, and image data is transmitted to a control processing hardware module 5;

[0136] S2, a real-time digital graphics processing chip 8 is used to perform geometric correction, size standardization, elevation angle distortion correction and fish tank area cutting on the image signal, specifically including:

[0137] S21, a scaling factor is dynamically calculated based on image width and height parameters, and the image is compressed to a preset size;

[0138] S22, the geometric transformation unit 13 is used to perform image rotation transformation, perspective correction and scaling operation;

[0139] S23, the camera intrinsic matrix is used to correct the perspective distortion of the elevation angle shooting picture;

[0140] S24, a rectangular cutting mask is generated by the mask generation unit 14 to remove the redundant background;

[0141] S3, the artificial intelligence processing unit 11 calls the model to perform multi-fish body detection, tracking and coordinate mapping to generate structured data containing fish body position coordinates and velocity vectors, specifically including:

[0142] S31, a deep separable convolution and cross-stage residual structure are used to construct a backbone network to output multi-scale feature maps;

[0143] S32, candidate boxes are generated according to anchor box parameters, and a dynamic threshold strategy is used in combination with a weighted NMS algorithm to screen target boxes;

[0144] S33, the IoU threshold value is dynamically adjusted by the target density matrix to suppress redundant detection boxes;

[0145] S34, a motion-external feature correlation tracker is used to realize multi-target ID management;

[0146] S4, the coordinate mapping unit (15) and the display space transformation unit (16) are used to convert the detection coordinates into a fish body position set in the background display coordinate system and a velocity vector group, specifically including:

[0147] S41, the external feature uses ResNet-18 to output a 128-dimensional vector, and the motion prediction uses Kalman filtering;

[0148] S42, compensating the scaling and cropping operations through batch matrix operations to restore the detection coordinates to the original resolution;

[0149] S43, mapping the fish body position set in the original coordinate system to the display coordinate system;

[0150] S5, dynamic rendering: According to the structured data and user instructions, the physical engine 17 calculates the dynamic effect parameters, generates a dynamic image control signal and outputs it to the background display screen 4, specifically including:

[0151] S51, reading the current background mode and fish position / speed data;

[0152] S52, calling the GPU-accelerated fluid dynamics model to simulate the multi-ripple superposition effect and calculate the diffusion parameter matrix based on the vertical velocity component of the fish body;

[0153] S53, calculating the emission parameters of the fireworks particle swarm based on the fish body velocity vector, and using spatial hashing to optimize the overlapping area calculation;

[0154] S54, allocating calculation accuracy based on the fish body importance score, generating a dynamic effect of end-to-end delay;

[0155] After receiving the control signal, S6 and the background display screen 4 optimize the pixel response speed through the display driver chip and display the dynamic effect to the corresponding position, ensuring that the fish movement and the virtual background effect are linked in real time and the visual alignment is accurate.

[0156] As a preferred implementation, step S3 of this embodiment further includes:

[0157] S31, perform feature extraction through a depthwise separable convolutional backbone network;

[0158] S32, Fusion of multi-scale features through feature pyramid network to improve small target detection capability;

[0159] S33, dynamic threshold screening: adaptively adjust the confidence threshold according to the target density;

[0160] S34, trajectory tracking: perform correlation tracking of fish bodies in consecutive frames based on the Kalman filter algorithm.

[0161] As a preferred implementation, step S5 of this embodiment further includes:

[0162] S51, ripple effect generation: calculating multi-ripple superposition interference parameters based on the vertical velocity component of the fish body;

[0163] S52, particle system control: Generate emission parameters of fireworks particle groups with different colors and densities according to the fish body velocity vector;

[0164] S53, physical engine simulation: real-time calculation of physical parameters of dynamic effects through GPU-accelerated physical engine.

[0165] As a preferred embodiment, the embodiment also includes a multi-aquarium linkage process implementation method, and the specific steps are:

[0166] S61, synchronizing the fish population coordinate set of each fish tank and the device ID in real time through the multi-device networking interface 12;

[0167] S62, when the first fish tank fish moves to the preset area, triggering the second fish tank background display screen to generate a cooperative dynamic effect;

[0168] S63, building a cross-device virtual ecological scene, combining the fish population movement trajectory of multiple fish tanks into a migration animation segment display;

[0169] S64, global coordinate mapping: uniformly mapping the fish position data of multiple fish tanks to a shared coordinate system;

[0170] S65, linkage effect rendering: generating a cross-device linkage dynamic background effect based on global coordinates.

[0171] As a preferred embodiment, the embodiment also includes an AI-generated background interface implementation method, and the specific steps are:

[0172] S71, receiving user voice instructions through the control screen 3, and converting them into text through a voice recognition chip;

[0173] S72, transmitting the text to the cloud StableDiffusion model through the wireless communication hardware unit 10; StableDiffusion model is a deep learning-based generative model, mainly used for image generation tasks.

[0174] S73, downloading the generated 1080p@60fps dynamic video stream and superimposing it with the local fish position data for output;

[0175] S74, identifying the scene type based on the current fish distribution and behavior pattern;

[0176] S75, sending a background generation request to the cloud server through the wireless communication hardware unit;

[0177] S76, receiving and caching the dynamic background materials generated by the cloud;

[0178] S77, dynamically switching the adaptive background materials according to the fish position.

[0179] As a preferred embodiment, the embodiment also includes a gesture interaction implementation method, and the specific steps are:

[0180] S81, capture user gesture image by gesture recognition camera 19;

[0181] S82, run MediaPipe algorithm by deep neural network processor 20 to output 21 hand joint coordinates;

[0182] S83, map gesture position (hx, hy) to dynamic effect trigger point, gesture speed positively correlates with effect intensity;

[0183] S84, gesture feature extraction: extract gesture features by deep neural network processor;

[0184] S85, action classification: identify waving, clicking, and sliding actions based on pre-trained gesture classification model;

[0185] S86, coordinate mapping: map gesture coordinates to background display coordinate system;

[0186] S87, effect triggering: generate and display corresponding dynamic effects based on recognized gesture actions and coordinate positions.

[0187] Embodiment Four

[0188] Please refer to Figure 1 、 Figure 2 、 Figure 3 and Figure 4 , the signal processing flow of fish body position detection and dynamic background linkage of the present embodiment includes the following steps:

[0189] Step S1: The miniature camera 2 uses a global shutter CMOS sensor to capture real-time images inside the fish tank at a resolution of 1080p@60fps, and transmits the image data to the control processing hardware module 5 through a USB3.0 high-speed data interface.

[0190] Step S2: Real-time digital graphics processing chip 8 performs geometric transformation and feature extraction on input video frames, specifically including:

[0191] S21: Image orientation correction and size standardization

[0192] Read the original image data through the video interface to obtain the image width and height parameters. Use rotation transformation to adjust the image orientation to the standard direction, and dynamically calculate the scaling factor according to the original image size to compress the image to the pre-set maximum size range, ensuring that the image data volume adapts to the algorithmic requirements of the subsequent artificial intelligence processing unit.

[0193] S22: Elevation angle shooting picture correction and fish tank area cropping

[0194] For the scene where the camera is installed below the front of the fish tank and needs to be shot at an angle, the picture correction is realized through the following operations: first, based on the camera internal parameter matrix and distortion coefficient, the original image of the upward shooting is corrected for perspective distortion, eliminating the stretching deformation of the tank edge caused by the upward tilt of the lens; then, according to the actual size of the fish tank and the installation position, the fish tank boundary is manually or automatically calibrated in the corrected image, a rectangular cutting area mask corresponding to the real tank is generated, the redundant background outside the mask is removed, and the extracted image area is accurately matched with the real fish tank space.

[0195] Step S3: The artificial intelligence processing unit 11 calls the optimized YOLOv5 deep learning model to perform the following operations on the NPU of RK3588:

[0196] S31: Multi-scale feature extraction and candidate box generation

[0197] The preprocessed video frame is input into the improved YOLOv5 network (optimized based on RK3588 NPU), a deep separable convolution (3x3 kernel) and a cross-stage residual structure are used to construct the backbone network, and 13x13 to 52x52 multi-scale feature maps are output. Through the feature pyramid network (FPP), different levels of features are fused to improve the detection ability of small targets in dense fish groups. Based on K-means clustering analysis (5000 training samples) combined with fish length-width ratio statistics, the anchor box parameters are set to [1.2, 3.5], [2.8, 6.9], and initial candidate boxes are generated at three detection layers, supporting the detection of more than 50 fish targets at the same time.

[0198] S32: Dynamic threshold detection optimization

[0199] "Classify (Sigmoid) and regress (CIoULoss) predictions are performed on the candidate boxes to generate target confidence and positioning parameters. A two-stage screening strategy is used: first, the confidence threshold is dynamically adjusted based on target density (low density 0.6 / high density 0.8), where the target density matrix calculation formula is: target density matrix = number of detection boxes / image area, the number of detection boxes is the total number of fish target boxes identified in the current frame, and the image area is the pixel value of the preprocessed effective detection area (unit: pixel²). Then, an improved weighted NMS algorithm (IoU threshold 0.5-0.7 dynamic adjustment) is applied to adaptively suppress redundant boxes by calculating the detection box density matrix, ensuring effective separation of overlapping fish bodies, and maintaining a detection recall rate of >95% in typical scenarios."

[0200] S33: Multi-target tracking and identity preservation

[0201] Build a tracker based on the association of motion-appearance dual features: 1) Appearance features use lightweight ResNet-18 (output 128-dimensional vector, NPUINT8 quantization) to calculate cosine similarity; 2) Motion prediction uses Kalman filter (state vector [x, y, w, h, vx, vy]) combined with Mahalanobis distance measurement; 3) ID management introduces an exponential decay mechanism (decay factor λ = 0.05), retains ID information for 30 frames (corresponding to 1 second @ 30fps) when the target is lost, and realizes continuous identity after occlusion through trajectory interpolation algorithm. Experiments show that the ID switching rate is <0.1 times / minute.

[0202] S34: Multi-resolution coordinate mapping and display space alignment

[0203] For the final confirmed multi-fish target frame set, the following operations are performed: first, restore the fish detection coordinates from the resolution after AI processing (such as 800x600) to the original image resolution (such as 1080p), compensate for the scaling and cropping operations in the preprocessing stage through batch matrix operations, and generate a multi-fish position set in the original image coordinate system ; Then based on the physical parameters (screen size, pixel density, DPI) of the display device, a coordinate transformation matrix is constructed to map the original image coordinates to the display device coordinate system, generating a multi-fish position set in the display coordinate system , where represents the normalized coordinates of the ith fish in the display space; At the same time, combining the position change amount between the two consecutive frames, the moving speed vector group of each fish body is calculated ; Finally, the position coordinate group, speed vector group and fish body ID are bound as a structured data block, which is synchronized to the physical engine in the S4 stage through the shared memory interface, realizing the real-time synchronous linkage of real fish school movement and virtual effect.

[0204] Step S4: Multi-fish dynamic background effect parameter calculation

[0205] Step S4: Microprocessor 7 based on the display coordinate group {(px1, py1), (px2, py2), …} and the moving speed vector group {(vx1, vy1), (vx2, vy2), …} of each fish body output by the S34 stage, combined with the currently activated background display mode (such as ripples, fireworks, etc.), calls the GPU-accelerated physical engine to calculate the dynamic effect parameters in parallel, including:

[0206] S41: Multi-fish data aggregation and mode judgment Microprocessor 7 obtains the real-time position coordinate group and speed vector group of all recognized fish bodies from the multi-target tracking module, and reads the currently activated background display mode from the display control module, providing basic data for subsequent batch physical simulation calculation.

[0207] S42: Physical engine batch parameter calculation performs corresponding calculation according to the current background mode: Ripple mode: weighted calculation is performed on the vertical velocity component of each fish to generate a ripple diffusion parameter matrix centered on the position of each fish, and a GPU-accelerated fluid dynamics model is used to simulate the superposition and interference effects of multiple ripples. Fireworks mode: independent fireworks launch parameters are calculated based on the velocity vector of each fish, and multiple groups of particles with different colors, densities, and lifespans are generated, and GPU parallel computing is used to realize the synchronous blooming effect of multiple fireworks.

[0208] S43: Multi-target performance optimization uses GPU parallel computing architecture to realize batch processing of fish parameters, spatial hash optimization of physical effects in overlapping areas, and dynamic adjustment of calculation precision distribution based on fish importance score.

[0209] Step S5: Signal conversion and high-speed output

[0210] Step S5: The microprocessor 7 integrates the calculated dynamic effect parameters with the original fish school video stream, converts them into dynamic image control signals in HDMI2.0 format, encodes them in YUV420 format using a dedicated video encoder, and outputs them at 1080p@60fps using an HDMI2.0 high-speed video interface, ensuring that the video stream and dynamic effect parameters are transmitted synchronously and the picture is smooth and free of lag.

[0211] Step S6: The background display screen 4 uses a 60Hz high refresh rate panel, and after receiving the control signal, it optimizes the pixel response speed through the display driver chip to display the dynamic effect to the corresponding position, and the delay from signal reception to display update is controlled to be ≤12ms, ensuring that the fish movement and virtual background effects (such as ripples and fireworks) are linked in real time and the visual alignment is accurate.

[0212] The entire signal processing flow is accelerated by a hardware pipeline (such as a dedicated image processor for parallel processing of acquisition, calculation, and encoding), multi-threaded software optimization (such as simultaneous execution of detection and rendering tasks), and timing calibration mechanisms (such as frame synchronization phase-locked loop technology), and the end-to-end delay from image acquisition to display update is controlled within 50 milliseconds, ensuring that the user perceives a smooth and natural interaction between the fish and the background effects without noticeable lag.

[0213] As an extension of the implementation, the device can realize the following extended functions:

[0214] 1. Gesture interaction process: The environment camera (mini camera 3) collects user gesture images at a resolution of 640x480 and a frame rate of 30fps, and transmits them to the control processing hardware module through the USB3.0 interface: the real-time digital graphics processing chip 8 runs the MediaPipe gesture recognition algorithm, outputs 21 hand joint coordinates, and generates gesture positions ((hx, hy)) in the display coordinate system after coordinate mapping; the microprocessor 7 calls the corresponding background effect function according to the gesture type (such as waving hands, making fists), for example: the waving action triggers the ripple effect, and the center of the ripple is ((hx, hy)), and the diffusion speed is positively correlated with the gesture waving amplitude.

[0215] AI-generated background process:

[0216] The user speaks "I want a tropical rainforest background" through the control screen 3, and the built-in voice recognition chip converts the voice into text "tropical rainforest background".

[0217] The microprocessor 7 sends the text to the cloud AI service through the Wi-Fi module, and the cloud returns the generated 1080p@60fps dynamic video stream.

[0218] The video stream is transmitted to the background display screen 4 through the HDMI interface, and is superimposed with the dynamic effects (such as ripples, fireworks) generated by the real-time fish position.

[0219] Multi-aquarium linkage process:

[0220] Fish tank A (device ID: 001) and fish tank B (device ID: 002) are bound through the cloud platform, and each main camera uploads fish coordinates to the server in real time;

[0221] When the fish in fish tank A moves to the right side of the screen, the coordinate data is synchronized to fish tank B through the cloud;

[0222] The microprocessor 7 of fish tank B generates water flow animation in the corresponding direction on the left side of the background display screen according to the received coordinates, creating a "fish migration" visual effect across the fish tanks.

[0223] Working principle and application scenario of the present application:

[0224] The user can select a preset scene (such as "underwater world", "starlit prairie") through the touch interface of the control screen 3, or trigger the effect through voice commands (such as "start interactive mode"). The control processing hardware module 5 calls the dynamic background material library in the storage unit 9, and synchronously updates the content of the background display screen 4.

[0225] As a smart home decoration, the present application can be installed in the living room or study. Parents can set the interactive mode of "fish chasing virtual bait" for children through the control screen 3. When the fish swims towards the bait icon, the background display screen 4 synchronously displays the splashing effect, enhancing the fun of parent-child interaction. At night, it can be switched to the "quiet starry sky" mode, and the background generates meteor animation with the gradual starlight and the fish swimming track, creating an immersive viewing experience.

[0226] In commercial application scenarios such as aquariums, pet stores or exhibition halls, the background display screen 4 of the present application can play virtual ecological scenes (such as coral reefs, tropical rainforests), and virtual creatures (such as turtles, jellyfish) in the background will make avoidance or following actions when the fish swims, forming a dynamic symbiotic visual effect, attracting the audience to stop. Merchants can also remotely update the background content of all display fish tanks through the wireless communication hardware unit 10 (Wi-Fi chip) of the control processing hardware module 5 to uniformly manage the display effect.

[0227] In the office environment, the present application can be used as a smart stress reliever. The artificial intelligence processing unit 11 analyzes the behavior patterns of the fish and automatically adjusts the background display content to create a visual effect that matches the current environment, improving the aesthetics and comfort of the office space.

[0228] The present application also supports connection to a smart home system through the wireless communication hardware unit 10 to realize linkage with other smart devices. For example, the brightness of the background display screen 4 can be automatically adjusted according to the indoor lighting brightness; or automatically switched to "silent mode" during a specific time period (such as a meeting time) to turn off the dynamic effect and only keep the basic viewing function.

[0229] Although the embodiments of the application have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limiting the application. Those skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the application.

Claims

1. A method for realizing dynamic interaction of a fish tank based on position detection, characterized by: The signal processing process for fish position detection and dynamic background linkage includes the following steps: S1, collects the fish tank video stream through a micro camera (2) with a polarization filter (6), and transmits the image data to the control processing hardware module (5); S2, performing geometric correction, size standardization, elevation distortion correction and fish tank area cropping on the image signal through a real-time digital graphics processing chip (8), specifically including: S21, dynamically calculating a scaling factor based on image width and height parameters, and compressing the image to a preset size; S22, performing image rotation transformation, perspective correction and scaling operations through the geometric transformation unit (13); S23, performing perspective distortion correction on the elevation-angle shot based on the camera intrinsic parameter matrix; S24, generating a rectangular cropping mask through the mask generation unit (14) to remove redundant background; S3, calling the model to perform multi-fish detection, tracking and coordinate mapping through the artificial intelligence processing unit (11), generating structured data including fish position coordinates and velocity vectors, specifically including: S31. Use depth-wise separable convolution and cross-stage residual structure to build a backbone network and output multi-scale feature maps; S32. Generate candidate frames based on anchor frame parameters, and use dynamic threshold strategy combined with weighted NMS algorithm to screen target frames; S33, dynamically adjust the IoU threshold through the target density matrix to suppress redundant detection boxes; S34, multi-target ID management through motion-appearance dual feature association tracker; S4, through the coordinate mapping unit (15) and the display space transformation unit (16), the detection coordinates are converted into the fish body position set in the background display coordinate system And velocity vector group, specifically including: S41, appearance features use ResNet-18 to output 128-dimensional vectors, and motion prediction uses Kalman filtering; S42, compensating the scaling and cropping operations through batch matrix operations to restore the detection coordinates to the original resolution; S43, mapping the fish body position set in the original coordinate system to the display coordinate system; S5, dynamic rendering: According to the structured data and user instructions, the dynamic effect parameters are calculated through the physical engine (17), and the dynamic image control signal is generated and output to the background display screen (4). S6. After receiving the control signal, the background display screen (4) optimizes the pixel response speed through the display driver chip and displays the dynamic effect to the corresponding position, ensuring that the fish body movement and the virtual background effect are linked in real time and the visual alignment is accurate; Step S5 specifically includes: S51, reading the current background mode and fish position / speed data; S52, calling the GPU-accelerated fluid dynamics model to simulate the multi-ripple superposition effect and calculate the diffusion parameter matrix based on the vertical velocity component of the fish body; S53, calculating the emission parameters of the fireworks particle swarm based on the fish body velocity vector, and using spatial hashing to optimize the overlapping area calculation; S54, allocating calculation accuracy based on the fish body importance score, generating a dynamic effect of end-to-end delay; Step S3 also includes: S35, perform feature extraction through a depth-wise separable convolutional backbone network; S36, Fusion of multi-scale features through feature pyramid network to improve small target detection capability; S37, dynamic threshold screening: adaptively adjust the confidence threshold according to the target density; S38, trajectory tracking: performing correlation tracking of fish bodies in consecutive frames based on the Kalman filter algorithm; Step S5 also includes: S55, ripple effect generation: calculating multi-ripple superposition interference parameters based on the vertical velocity component of the fish body; S56, particle system control: Generate emission parameters of fireworks particle groups with different colors and densities according to the fish body velocity vector; S57, Physics Engine Simulation: Real-time calculation of physical parameters of dynamic effects through GPU-accelerated physics engine.

2. The method according to claim 1, wherein: It also includes the implementation method of the multi-fish tank linkage process, the specific steps are: S61. Synchronize the fish school coordinates and device IDs of each fish tank in real time through the multi-device networking interface; S62: When the fish in the first fish tank moves to a preset area, the background display screen of the second fish tank is triggered to generate a coordinated dynamic effect; S63, constructing a cross-device virtual ecological scene, synthesizing the movement trajectories of fish schools in multiple fish tanks into a segmented migration animation display; S64, global coordinate mapping: uniformly mapping the fish position data of multiple fish tanks to a shared coordinate system; S65. Linkage effect rendering: Generate dynamic background effects that are linked across devices based on global coordinates.

3. The method according to claim 2, wherein: It also includes the implementation method of AI-generated background interface, the specific steps are: S71, receiving user voice commands through the control screen (3), and converting them into text through the voice recognition chip; S72, transmitting the text to the cloud via the wireless communication unit (10); S73, downloading the generated dynamic video stream, and superimposing it with the local fish position data for output; S74, identifying the scene type based on the current fish distribution and behavior pattern; S75, sending a background generation request to the cloud server via the wireless communication unit; S76. Receive and cache dynamic background materials generated in the cloud; S77. Dynamically switch the appropriate background material according to the position of the fish.

4. The method according to claim 3, wherein: It also includes the implementation method of gesture interaction, the specific steps are: S81, collecting user gesture images through a gesture recognition camera; S82, deep neural network processor outputs the coordinates of hand joints; S83, mapping the gesture position (hx, hy) to the dynamic effect trigger point, where the gesture speed is positively correlated with the effect intensity; S84, gesture feature extraction: extracting gesture features through a deep neural network processor; S85, Action Classification: Recognize waving, clicking, and sliding actions based on a pre-trained gesture classification model; S86, coordinate mapping: mapping the gesture coordinates to the background display coordinate system; S87, effect triggering: generating and displaying corresponding dynamic effects based on the recognized gesture action and coordinate position.

5. The method according to any one of claims 1 to 4, characterized in that: A fish tank dynamic interactive display device based on position detection, comprising: Fish tank body (1); A miniature camera (2) is used to capture images of fish in the fish tank in real time and output image signals. A polarizing filter (6) is provided in front of the lens of the miniature camera (2) to shield light interference from the background display screen (4); The control screen (3) is a touch screen hardware unit with an integrated microphone, used to display system status and receive user instructions; A background display screen (4) is installed behind the fish tank body (1) and is used to play a dynamic background and display dynamic effects according to the position of the fish; A control processing hardware module (5) includes a microprocessor (7), a real-time digital graphics processing chip (8), an artificial intelligence processing unit (11) and a storage unit (9), wherein the microprocessor (7) is connected to the micro camera (2), the control screen (3) and the background display screen (4), and is used to: Processing the image signal by the real-time digital graphics processing chip (8) to obtain the position information of the fish; Control background display content according to user instructions; The fish position information is converted into an image control signal containing dynamic effect parameters and output to the background display screen (4), wherein the real-time digital graphics processing chip (8) integrates a geometric transformation unit (13) and a mask generation unit (14), which are respectively used for image correction and fish tank area cropping; the artificial intelligence processing unit (11) is used to perform multi-fish body detection, tracking and coordinate mapping; the control processing hardware module (5) further includes a coordinate mapping unit (15) and a display space transformation unit (16), which are used to map the fish body coordinates to the display coordinate system, and integrates a physical engine (17) for calculating the dynamic background effect parameters based on the fish body position.

6. The device according to claim 5, characterized in that The system further includes a gesture interaction module (18), wherein the gesture interaction module (18) includes a gesture recognition camera (19) and a deep neural network processor (20), wherein the gesture recognition camera (19) is used to collect gesture images, and the deep neural network processor (20) is used to recognize gestures and trigger dynamic effects.

7. The device according to claim 6, characterized in that The control processing hardware module (5) further includes a multi-device networking interface (12) that supports synchronization of fish position data of multiple fish tanks via a local area network, thereby realizing cross-device background linkage.

Images