Image acquisition system under dynamic ambient light, ambient light encoding method and device
By using a spherical architecture and ambient lighting encoding method, and by providing preset RGB channel values from multiple fixed light sources, the problem of low efficiency and realism in capturing dynamic object images under dynamic ambient lighting is solved, achieving efficient and realistic image capture results.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- QINGDAO TAPER ROBOTICS CO LTD
- Filing Date
- 2026-02-27
- Publication Date
- 2026-05-19
AI Technical Summary
Existing technologies struggle to efficiently and realistically capture images of dynamic objects under dynamic ambient lighting, especially when facial expressions change. This leads to difficulties in image alignment, low production efficiency, and difficulty in achieving physical consistency with complex and realistic lighting conditions.
The image acquisition system adopts a spherical architecture, which uses multiple uniformly distributed fixed light sources to provide independent preset RGB channel values, and controls the switching of light sources and the shooting synchronization through an ambient light coding method to achieve image acquisition under dynamic ambient light.
It enables efficient and realistic image acquisition of dynamic objects under dynamic ambient lighting conditions, overcoming the limitations of traditional methods. It can quickly switch lighting conditions and is suitable for image acquisition of dynamic objects.
Smart Images

Figure CN121751002B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of image acquisition technology, and in particular to an image acquisition system, an ambient lighting encoding method, and an apparatus for dynamic ambient lighting. Background Technology
[0002] In industries such as film and television special effects and mobile phone imaging, dynamic ambient lighting is often applied to the subject to blend it with different scenes, so that the lighting effects of the subject can realistically match the background.
[0003] According to relevant technologies, realistic composite images are often generated by manually editing the materials and textures of the photographed object, arranging virtual scenes and light sources in 3D software, and drawing based on rendering equations. However, this method is highly dependent on the experience of artists and repeated adjustments, resulting in low production efficiency and difficulty in achieving physical consistency with complex and realistic lighting. Alternatively, high dynamic range ambient light maps of specific environments can be collected, and image-based lighting techniques can be used to achieve high-quality relighting of virtual or real objects. However, for dynamic objects (such as human faces), long-term shooting can lead to changes in posture and expression, causing difficulties in subsequent image alignment.
[0004] Therefore, finding a system that can efficiently and realistically acquire images of dynamic objects under dynamic ambient lighting has become a current research hotspot. Summary of the Invention
[0005] This invention provides an image acquisition system, an ambient lighting encoding method, and an apparatus for dynamic ambient lighting, enabling efficient, realistic image acquisition of dynamic objects under dynamic ambient lighting conditions.
[0006] This invention provides an image acquisition system under dynamic ambient lighting. The system includes: a spherical structure, with a target object placed at the center of the sphere; multiple fixed light sources evenly distributed on the surface of the sphere, each fixed light source independently providing a light source with preset RGB channel values for the target object; a control module for sending light source control signals to the fixed light sources, so that the fixed light sources, upon receiving the light source control signals, independently provide a light source with preset RGB channel values for the target object, wherein the light source control signals are implemented using an ambient lighting encoding method; and a shooting module, the shooting direction of which is directed towards the center of the sphere, for shooting the target object under the condition that each fixed light source independently provides a light source with preset RGB channel values for the target object, to obtain an image of the target object under preset ambient lighting, wherein the preset ambient lighting is formed by each fixed light source independently providing a light source with preset RGB channel values for the target object.
[0007] According to the image acquisition system under dynamic ambient lighting provided by the present invention, the control module further includes: a synchronization control module, used to send synchronization control signals to the fixed light source and the imaging module, so that the light source switching of the fixed light source and the image acquisition of the imaging module are performed synchronously.
[0008] According to the present invention, an image acquisition system under dynamic ambient lighting is provided, wherein the spherical structure includes a regular 32-dodecahedron, the regular 32-dodecahedron includes multiple vertices and multiple midpoints of edges; the fixed light source is respectively disposed at the vertex and / or the midpoint of the edge.
[0009] According to the present invention, an image acquisition system under dynamic ambient lighting conditions is provided, wherein the fixed light source is composed of an LED bead matrix; the LED bead matrix is controlled by binary encoding; and the LED bead matrix has preset RGB channel values.
[0010] According to the image acquisition system under dynamic ambient lighting provided by the present invention, the fixed light source is further configured with a light source address; the control module sends a light source control signal to the fixed light source through the light source address.
[0011] This invention also provides an ambient lighting encoding method, the method comprising: using the center of a spherical structure as the origin, calibrating the positions of multiple fixed light sources within the spherical structure and performing normalization processing to obtain a direction matrix matching the positions of the fixed light sources, wherein the elements in the direction matrix are used to characterize the light source direction of the fixed light source; determining a distance vector based on the direction matrix, wherein the distance vector is used to characterize the light source distribution density of the fixed light source; setting a kurtosis parameter of a spherical Gaussian function based on the distance vector; traversing all fixed light sources, and for any fixed light source, performing the following steps: determining a spherical Gaussian function based on the light source direction of the fixed light source and the kurtosis parameter. Sampling kernel; based on the spherical Gaussian sampling kernel, the target dynamic range ambient light map is sampled to obtain RGB channel values that match the target dynamic range ambient light map corresponding to the fixed light source, wherein the target dynamic range ambient light map is a light map that matches the image of the object to be photographed under a preset ambient light; the RGB channel values corresponding to each of the fixed light sources are combined to obtain a combined RGB channel value sequence; based on the combined RGB channel value sequence, encoding processing is performed to obtain a binary code of a light source with preset RGB channel values that matches the target dynamic range ambient light map, and a light source control signal is obtained based on the binary code.
[0012] According to an ambient lighting encoding method provided by the present invention, the step of determining a distance vector based on the direction matrix includes: determining a distance matrix between different light sources based on the direction matrix, wherein the elements in the distance matrix are used to characterize the distance between different fixed light sources; taking the largest matrix column element along each matrix column of the distance matrix, and / or taking the largest matrix row element along each matrix row of the distance matrix; and determining the distance vector based on the largest matrix column element and / or the largest matrix row element.
[0013] According to an ambient lighting encoding method provided by the present invention, the step of sampling a target dynamic range ambient light map based on the spherical Gaussian sampling kernel to obtain RGB channel values corresponding to the fixed light source and matching the target dynamic range ambient light map includes: sampling the target dynamic range ambient light map based on the spherical Gaussian sampling kernel to obtain multiple channel values under each RGB channel; averaging the multiple channel values under each RGB channel to obtain the channel average value under each channel; and obtaining the RGB channel values corresponding to the fixed light source and matching the target dynamic range ambient light map based on the channel average value under each channel.
[0014] This invention also provides a method for image acquisition under dynamic ambient lighting, the method comprising: encoding a target dynamic range ambient light map into a light source control signal with preset RGB channel values that matches the target dynamic range ambient light map, according to any one of the ambient lighting encoding methods described herein; controlling fixed light sources to switch lighting states based on the light source control signal, so that each fixed light source independently provides a light source with preset RGB channel values for the object to be photographed; and triggering a shooting module to photograph the object to be photographed when each fixed light source independently provides a light source with preset RGB channel values for the object to be photographed, thereby obtaining an image of the object to be photographed under preset ambient lighting.
[0015] According to a dynamic ambient lighting image acquisition method provided by the present invention, the target dynamic range ambient light map includes multiple target dynamic range ambient light maps; the method of encoding the target dynamic range ambient light map into a light source control signal with preset RGB channel values matching the target dynamic range ambient light map according to any one of the ambient lighting encoding methods includes: encoding multiple target dynamic range ambient light maps into a sequence of light source control signals with preset RGB channel values matching the multiple target dynamic range ambient light maps according to any one of the ambient lighting encoding methods; and controlling fixed light sources to switch lighting states based on the light source control signals so that each fixed light source independently provides a light source with preset RGB channel values for the object to be photographed, including: based on the light source... A control signal sequence controls fixed light sources to switch illumination states, so that each fixed light source independently provides light with preset RGB channel values to the object to be photographed according to the order of the light source control signal sequence. The step of triggering the shooting module to photograph the object under preset ambient lighting, when each fixed light source independently provides light with preset RGB channel values to the object, to obtain an image of the object under preset ambient lighting, includes: triggering the shooting module to photograph the object under preset ambient lighting, when each fixed light source independently provides light with preset RGB channel values to the object according to the order of the light source control signal sequence, to obtain an image sequence of the object under preset ambient lighting; and combining the image sequence of the object under preset ambient lighting to obtain a video clip.
[0016] The present invention also provides an ambient lighting encoding device, the device comprising: a calibration module, configured to calibrate the positions of multiple fixed light sources in a spherical structure with the center of the sphere as the origin and perform normalization processing to obtain a direction matrix matching the positions of the fixed light sources, wherein the elements in the direction matrix are used to characterize the light source direction of the fixed light source; a determination module, configured to determine a distance vector based on the direction matrix, wherein the distance vector is used to characterize the light source distribution density of the fixed light source; a setting module, configured to set the kurtosis parameter of the spherical Gaussian function based on the distance vector; and a processing module, configured to traverse all fixed light sources and, for any fixed light source, perform the following steps: based on the light source direction of the fixed light source and the kurtosis parameter. The system comprises: determining a spherical Gaussian sampling kernel; sampling the target dynamic range ambient light map based on the spherical Gaussian sampling kernel to obtain RGB channel values that match the target dynamic range ambient light map corresponding to the fixed light source, wherein the target dynamic range ambient light map is a light map that matches the image of the object to be photographed under a preset ambient light; a combination module is used to combine the RGB channel values corresponding to each of the fixed light sources to obtain a combined RGB channel value sequence; and a generation module is used to encode the combined RGB channel value sequence to obtain a binary code of a light source with preset RGB channel values that matches the target dynamic range ambient light map, and to obtain a light source control signal based on the binary code.
[0017] The present invention also provides an image acquisition device under dynamic ambient lighting, the device comprising: an encoding module, configured to encode a target dynamic range ambient light map into a light source control signal having preset RGB channel values that matches the target dynamic range ambient light map according to any one of the ambient lighting encoding methods described herein; a switching module, configured to control fixed light sources to switch lighting states based on the light source control signal, so that each fixed light source independently provides a light source with preset RGB channel values for the object to be photographed; and an acquisition module, configured to trigger an acquisition module to photograph the object to be photographed when each fixed light source independently provides a light source with preset RGB channel values for the object to be photographed, thereby obtaining an image of the object to be photographed under preset ambient lighting.
[0018] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any of the above-described ambient lighting encoding methods, or to execute an image acquisition method under dynamic ambient lighting.
[0019] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the ambient lighting encoding method as described above, or performs an image acquisition method under dynamic ambient lighting.
[0020] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements any of the above-described ambient lighting encoding methods, or performs an image acquisition method under dynamic ambient lighting.
[0021] This invention provides an image acquisition system, ambient lighting encoding method, and apparatus for dynamic ambient lighting. The system includes multiple fixed light sources, a shooting module, and a control module. The object to be photographed is placed at the center of a spherical structure. Multiple fixed light sources are evenly distributed on the surface of the spherical structure, each providing an independent light source with preset RGB channel values for the object. The control module sends light source control signals to the fixed light sources, enabling each fixed light source to independently provide an image with preset RGB channel values for the object upon receiving the control signals. The shooting module shoots towards the center of the spherical structure, capturing images of the object under preset ambient lighting conditions when each fixed light source independently provides an image with preset RGB channel values. This invention enables efficient, realistic, and applicable image acquisition of dynamic objects under dynamic ambient lighting conditions, thereby obtaining an image of the object under preset ambient lighting conditions. Attached Figure Description
[0022] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0023] Figure 1 This is a schematic diagram of the hardware environment for an ambient lighting encoding method or an image acquisition method under dynamic ambient lighting, according to an embodiment of this application.
[0024] Figure 2 This is a schematic diagram of the image acquisition system under dynamic ambient lighting provided by the present invention.
[0025] Figure 3 This is a flowchart illustrating the ambient lighting encoding method provided by the present invention.
[0026] Figure 4 This is a flowchart illustrating the image acquisition method under dynamic ambient lighting provided by the present invention.
[0027] Figure 5 This is a schematic diagram of the ambient light encoding device provided by the present invention.
[0028] Figure 6 This is a schematic diagram of the image acquisition device under dynamic ambient lighting provided by the present invention.
[0029] Figure 7 This is a schematic diagram of the electronic device provided by the present invention.
[0030] Figure label:
[0031] 102: Terminal equipment; 104: Server; 200: Image acquisition system under dynamic ambient lighting; 210: Spherical architecture; 220: Fixed light source; 230: Control module; 240: Imaging module. Detailed Implementation
[0032] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.
[0033] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0034] According to one aspect of the embodiments of this application, an ambient light encoding method or an image acquisition method under dynamic ambient light is provided. This ambient light encoding method or image acquisition method under dynamic ambient light is widely used in whole-house intelligent digital control application scenarios such as smart homes, smart home ecosystems, and intelligence house ecosystems. Optionally, in this embodiment, the above-mentioned ambient light encoding method or image acquisition method under dynamic ambient light can be applied to, for example... Figure 1The hardware environment shown consists of terminal device 102 and server 104. For example... Figure 1 As shown, server 104 is connected to terminal device 102 via a network and can be used to provide services (such as application services) to the terminal or clients installed on the terminal. A database can be set up on the server or independently of the server to provide data storage services for server 104. Cloud computing and / or edge computing services can be configured on the server or independently of the server to provide data processing services for server 104.
[0035] The aforementioned network may include, but is not limited to, at least one of the following: wired network, wireless network. The aforementioned wired network may include, but is not limited to, at least one of the following: wide area network, metropolitan area network, local area network. The aforementioned wireless network may include, but is not limited to, at least one of the following: Wi-Fi (Wireless Fidelity), Bluetooth. The terminal device 102 may not be limited to PC, mobile phone, tablet computer, smart air conditioner, smart range hood, smart refrigerator, smart oven, smart stove, smart washing machine, smart water heater, smart washing equipment, smart dishwasher, smart projector, smart TV, smart clothes rack, smart curtains, smart audio-visual equipment, smart socket, smart speaker, smart speaker box, smart fresh air equipment, smart kitchen and bathroom equipment, smart bathroom equipment, smart robot vacuum cleaner, smart window cleaning robot, smart mopping robot, smart air purifier, smart steam oven, smart microwave oven, smart water heater, smart air purifier, smart water dispenser, smart door lock, etc.
[0036] In another embodiment, the ambient lighting encoding method or the image acquisition method under dynamic ambient lighting provided in this application can be applied to smart home appliances. Smart home appliances refer to home appliance products formed by introducing microprocessors, sensor technology, and network communication technology into home appliance equipment. They have the ability to automatically sense the status of the residential space, the status of the appliances themselves, and the service status of the appliances, and can automatically control and receive control commands from the home user inside the residence or remotely. It is understood that smart home appliances are a component of smart homes.
[0037] Figure 2 This is a schematic diagram of the image acquisition system under dynamic ambient lighting provided by the present invention.
[0038] The following will combine Figure 2 The structure of the image acquisition system under dynamic ambient lighting provided by the present invention will be described.
[0039] In an exemplary embodiment of the present invention, combined with Figure 2 As can be seen, the image acquisition system 200 under dynamic ambient lighting may include a spherical structure 210, multiple fixed light sources 220, a control module 230, and an image capture module 240. Each module will be described in detail below.
[0040] The object to be photographed is placed at the center of the sphere in the spherical structure 210.
[0041] Multiple fixed light sources 220 are evenly distributed on the spherical surface of the spherical structure 210, and each fixed light source 220 independently provides a light source with preset RGB channel values for the object to be photographed;
[0042] Control module 230 is used to send a light source control signal to fixed light source 220 so that fixed light source 220 independently provides light source with preset RGB channel values for the subject to be photographed after receiving the light source control signal. The light source control signal is implemented using the ambient light coding method described below.
[0043] The shooting direction of the shooting module 240 is pointed to the center of the sphere of the spherical structure 210. It is used to shoot the subject under the condition that each fixed light source 220 independently provides light sources with preset RGB channel values to the subject, so as to obtain an image of the subject under preset ambient light. The preset ambient light is formed by each fixed light source 220 independently providing light sources with preset RGB channel values to the subject.
[0044] In one embodiment, the spherical architecture 210 can be a near-spherical physical support structure, with its internal space used to accommodate the object to be photographed, such as a person or still life. The geometric center of this architecture is explicitly defined as the center of the sphere, and the arrangement of all other components is based on this center. In one example, the spherical architecture 210 can be constructed from a polyhedral frame formed by subdividing a regular 32-dodecahedron to provide uniform and stable light source mounting points.
[0045] In another embodiment, the fixed light sources 220 can be installed in a high-density and uniform manner on the inner spherical surface of the spherical structure 210. Taking the spherical structure 210 described above as being formed by subdividing a regular 32-sided polyhedron as an example, the corresponding number of fixed light sources 220 can be 356. Each fixed light source 220 is an independently controllable light-emitting unit, the core of which is an RGB LED module. This module can receive digital control signals and accurately output light with specific brightness values for the red (R), green (G), and blue (B) channels. The light from all light sources points towards the center of the sphere, collectively providing illumination for the object to be photographed located at the center of the sphere.
[0046] In another embodiment, the control module 230 can be considered the "brain" of the system, typically consisting of a computer and dedicated light source controller hardware. Its core function is to generate and send independent light source control signals to each fixed light source 220. This signal is a digital instruction packet containing the address code of the target light source and its corresponding preset RGB channel values. When a fixed light source 220 receives its control signal, its internal drive circuit immediately adjusts the current to make the LED emit light completely consistent with the signal instruction.
[0047] In another embodiment, the imaging module 240 may include one or more high-resolution digital cameras. These cameras are fixed at specific locations on the spherical structure 210, such as at an opening, and are precisely calibrated to ensure that their lens optical axes are precisely aligned with the center of the sphere. The cameras can be connected to the imaging module 240 via data cables and can receive shooting commands.
[0048] During application, the object to be photographed can be placed at the center of the spherical structure 210. The operator sets a desired preset ambient lighting scheme using control software. This scheme is essentially a set of data defining the RGB values that each fixed light source 220 on the sphere should emit at this time. The control module 230 generates and broadcasts corresponding light source control signals based on this scheme. All fixed light sources 220 are simultaneously illuminated, and their emitted light superimposes on the surface of the object at the center of the sphere, collectively forming the required preset ambient lighting conditions. The control module 230 triggers the shooting module 240, and the camera captures an image of the object at this moment. Thus, an image of the object under the preset ambient lighting is obtained.
[0049] In another embodiment, the control module 230 can pre-store multiple different lighting schemes (corresponding to multiple different ambient light map encoding results) and cyclically send the corresponding control signals at the video frame rate, while simultaneously triggering the camera to continuously capture images. This enables efficient acquisition of image sequences or videos of the same object under various different and rapidly changing ambient lighting conditions.
[0050] In this embodiment, the system can physically synthesize arbitrarily complex spatial lighting distributions using a spherical array composed of hundreds of independent RGB light sources. This overcomes the limitation of traditional studio lighting, which cannot simulate the complex ambient lighting in the real world or for specific artistic effects, thus enabling the acquisition of images or image sequences of the subject under difficult-to-arrange ambient lighting conditions.
[0051] This invention provides an image acquisition system under dynamic ambient lighting, comprising: a spherical structure, multiple fixed light sources, a shooting module, and a control module. The object to be photographed is placed at the center of the sphere of the spherical structure. Multiple fixed light sources are evenly distributed on the surface of the spherical structure, each providing an independent light source with preset RGB channel values for the object. The control module sends light source control signals to the fixed light sources, enabling each fixed light source to independently provide an image with preset RGB channel values for the object upon receiving the control signals. The shooting module shoots towards the center of the sphere of the spherical structure, capturing images of the object under preset ambient lighting conditions when each fixed light source independently provides an image with preset RGB channel values. This invention enables efficient, realistic, and applicable image acquisition of dynamic objects under dynamic ambient lighting conditions, thereby obtaining an image of the object under preset ambient lighting conditions.
[0052] In yet another exemplary embodiment of the present invention, continuing with Figure 2 The above embodiment is used as an example for illustration. The control module 230 may also include a synchronization control module. The synchronization control module is used to send synchronization control signals to the fixed light source 220 and the imaging module 240 so that the light source switching of the fixed light source 220 and the image acquisition of the imaging module 240 are performed synchronously.
[0053] In one embodiment, the synchronization control module can be composed of a hardware signal generator, such as a crystal clock circuit and corresponding software driver. During application, when the image acquisition system 100 starts an acquisition process under dynamic ambient lighting, the control module 230 first sends a preset RGB data packet of the light source to the buffer of each fixed light source 220. Subsequently, the synchronization control module generates a high-precision synchronization control signal, such as a pulse signal with a defined rising or falling edge. This synchronization control signal is simultaneously broadcast to all fixed light sources 220 and the imaging module 240. For the fixed light sources 220, this signal serves as an execution command, triggering all light sources to switch to a new RGB illumination state at the same time, based on the received buffer data. For the imaging module 240, this signal serves as a shutter trigger command, driving the camera to perform exposure and capture at the same moment the light source state switching is completed. Through this mechanism, the switching action of the light source state and the image acquisition action of the camera are forcibly bound to the same time base, achieving hardware-level synchronous execution.
[0054] In another exemplary embodiment of the present invention, the spherical structure 210 may include a regular 32-sided polyhedron, which includes multiple vertices and multiple midpoints of edges; fixed light sources 220 are respectively disposed at the vertices and / or midpoints of edges.
[0055] In one embodiment, the spherical structure 210 may include a regular treble dodecahedron, wherein the regular treble dodecahedron can be a regular polyhedron consisting of 32 regular polygonal faces, 92 vertices, and 270 edges. This geometry is chosen because its high symmetry provides a uniform reference grid for the distribution of light sources that approximates an ideal sphere.
[0056] Furthermore, to achieve high-density and uniform coverage of the light source on the sphere, the mounting points of the fixed light source 220 can be directly associated with the geometric feature points of the regular dodecahedron. Specifically, the mounting points are selected at the vertices and / or the midpoints of the edges of the regular dodecahedron.
[0057] In one example, mounting points are placed at all 92 vertices and the midpoints of all 270 edges. Theoretically, this provides 92 + 270 = 362 potential light source placement locations. In actual physical implementation, to allow for personnel access, equipment installation, and camera viewing, light sources can be selectively left unmounted at a few points (e.g., 6). Therefore, the actual number of fixed light sources can be 356, evenly and symmetrically distributed on a spherical surface defined by the vertices and midpoints of the edges of a regular dodecahedron.
[0058] In another exemplary embodiment of the present invention, the fixed light source 220 may be composed of an LED bead matrix; the LED bead matrix is controlled by binary encoding; the LED bead matrix has preset RGB channel values.
[0059] In one embodiment, the light-emitting core of each fixed light source 220 can be a miniature module (dot matrix) integrating multiple LED beads. In another embodiment, the dot matrix is further designed as two concentric LED arrays, an inner ring and an outer ring, to improve the angular uniformity of light emission. The LED beads in the dot matrix are grouped by color, including at least red, green, and blue primary color beads. To enhance brightness and color adjustment range, white beads can also be added. Therefore, a complete dot matrix can be composed of a combination of four-color LED beads.
[0060] Color and brightness control for each fixed light source 220 is achieved through binary encoding. Specifically, an 8-bit binary code can be assigned to the brightness of each color. The light source control signal generated by the control module 230 is a data packet containing the address of the light source and the 8-bit encoded values of each color channel. Each fixed light source 220 has a decoding unit within its driving circuit. When the light source receives the control signal, the decoding unit confirms the instruction based on its own address code in the signal, parses the corresponding 8-bit binary value for each color, and then converts it into a precise driving current, causing each color LED to emit light at the specified brightness. After all the colors of light are mixed, a light source with preset RGB channel values is generated. Hundreds of such independently addressable, binary-encoded LED dot matrices are evenly installed on a spherical architecture based on a regular 32-sided polyhedron, together forming a programmable spherical lighting system.
[0061] In another exemplary embodiment of the present invention, the fixed light source 220 may also be configured with a light source address; the control module 230 sends the light source control signal to the fixed light source 220 through the light source address.
[0062] In one embodiment, each fixed light source 220 is configured with a unique light source address, which is a key identifier for achieving independent addressing and control of the light sources in the hardware system. During application, when it is necessary to update the state of all light sources, the control module 230 generates a series of data frames containing different addresses and corresponding RGB data in a predetermined order or according to the needs of the scene, and sends them out through the bus. All light sources receive the data in parallel, but each only responds to its own instructions, thereby achieving parallel, independent, and precise assignment of values to hundreds of light sources.
[0063] This invention provides an image acquisition system under dynamic ambient lighting, comprising: multiple fixed light sources, an imaging module, and a control module. The object to be photographed is placed at the center of a spherical structure. Multiple fixed light sources are evenly distributed on the surface of the spherical structure, each providing an independent light source with preset RGB channel values for the object. The control module sends light source control signals to the fixed light sources, enabling each fixed light source to independently provide an image with preset RGB channel values for the object upon receiving the control signals. The imaging module's shooting direction is pointed towards the center of the spherical structure, and it is used to photograph the object while each fixed light source independently provides an image with preset RGB channel values, thereby obtaining an image of the object under preset ambient lighting. This invention enables efficient, realistic, and applicable image acquisition of dynamic objects under dynamic ambient lighting, thus obtaining an image of the object under preset ambient lighting.
[0064] Based on the same inventive concept, the present invention also provides an ambient lighting encoding method. Figure 3This is a flowchart illustrating the ambient lighting encoding method provided by the present invention. The following will be combined with... Figure 3 The process of the ambient lighting encoding method provided by the present invention will be described.
[0065] In an exemplary embodiment of the present invention, the ambient lighting encoding method can be applied to any of the dynamic ambient lighting image acquisition systems described herein. The purpose of this method is to efficiently and accurately encode an arbitrary given target high dynamic range (HDR) ambient light map into light source control signals that drive all fixed light sources on a spherical architecture. Combined with... Figure 3 As can be seen, the ambient lighting encoding method may include steps 310 to 360, and each step will be described below.
[0066] In step 310, the positions of multiple fixed light sources in the spherical structure are calibrated and normalized using the center of the sphere as the origin, to obtain a direction matrix that matches the position of the fixed light sources. The elements in the direction matrix are used to characterize the light source direction of the fixed light sources.
[0067] In one embodiment, the center of the sphere in the spherical structure is used as the origin of the three-dimensional coordinate system. Based on the system's physical design, such as a regular dodecahedron, the three-dimensional coordinates of each fixed light source are accurately measured or calculated. All coordinates are then normalized by dividing by the modulus to obtain the direction vector of each light source on a unit sphere. Finally, the direction vectors of all n light sources are arranged in order to form a direction matrix. , among which, the The line represents the first The light source direction is fixed at a point.
[0068] In step 320, the distance vector is determined based on the direction matrix, wherein the distance vector is used to characterize the light source distribution density of the fixed light source.
[0069] In one embodiment, the direction matrix can be used as a basis. Calculate the distance vector reflecting the relative positional relationship between light sources. This can be achieved through the following steps: based on the direction matrix... Calculate the angular distance or Euclidean distance between any two point light sources and construct the distance matrix between the light sources. ,in, express An identity matrix of order 1. Then, along the distance matrix between light sources... The maximum value of a row or column is taken to obtain the distance vector representing the light source distribution density. This vector visually represents the density of light sources in different regions of the sphere; the smaller the value, the denser the light sources are.
[0070] In step 330, the kurtosis parameter of the spherical Gaussian function is set based on the distance vector.
[0071] In one embodiment, the distance vector can be used. Set the spherical Gaussian function steepness parameter Specifically, ,in, This represents d in the i-th row; Let d represent the mean of vector d.
[0072] In step 340, all fixed light sources are traversed, and for any fixed light source, the following steps are performed: a spherical Gaussian sampling kernel is determined based on the light source direction and steepness parameters of the fixed light source; based on the spherical Gaussian sampling kernel, the target dynamic range ambient light map is sampled to obtain the RGB channel values of the matching target dynamic range ambient light map corresponding to the fixed light source, wherein the target dynamic range ambient light map is the light map that matches the image of the object to be photographed under the preset ambient light.
[0073] In one embodiment, all n fixed light sources are traversed. For the i-th light source, the following steps can be performed:
[0074] Based on its light source direction and steepness parameter Construct its dedicated spherical Gaussian sampling kernel For example, for the i-th point light source, its direction vector can be used. Centered on, with To determine the sharpness parameter, construct a spherical Gaussian sampling kernel. ,in, Let be a unit vector in any direction on the surface of the sphere;
[0075] Using this sampling kernel, the input target high dynamic range ambient light map (img) is weighted and sampled, and the weighted average value is calculated for each color channel to obtain the target color value corresponding to the point light source. That is, to obtain the RGB channel values of the target dynamic range ambient light map that corresponds to the fixed light source.
[0076] In step 350, the RGB channel values corresponding to each fixed light source are combined to obtain a combined RGB channel value sequence.
[0077] In step 360, encoding processing is performed based on the combined RGB channel value sequence to obtain the binary code of the light source with preset RGB channel values that matches the target dynamic range ambient light map, and the light source control signal is obtained based on the binary code.
[0078] In one embodiment, traversing all From a point light source, obtain a complete color mapping set. This process involves obtaining a sequence of combined RGB channel values. Then, based on this sequence, encoding is performed to obtain the binary code of a light source with preset RGB channel values that matches the target dynamic range ambient ray map. Finally, the light source control signal is obtained based on this binary code. For example, a complete color map set can be converted into a binary control instruction file compatible with the light source hardware interface, thus obtaining binary code, which in turn leads to the light source control signal.
[0079] In this embodiment, by introducing a spherical Gaussian function as a mathematical bridge, this method can intelligently calculate the light color contribution of a finite number of discrete point light sources from a complete and continuous high dynamic range ambient occlusion map. This solves the fundamental problem of how to simulate continuous illumination in infinite directions using hundreds of point light sources, resulting in high decoding fidelity.
[0080] In yet another exemplary embodiment of the present invention, continuing with the previously described embodiment as an example, the distance vector can be determined based on the direction matrix in the following manner:
[0081] Based on the direction matrix, a distance matrix between different light sources is determined, wherein the elements in the distance matrix are used to characterize the distance between different fixed light sources;
[0082] Take the largest column element of each column along the distance matrix, and / or take the largest row element of each row along the distance matrix;
[0083] The distance vector is determined based on the largest column elements and / or the largest row elements of the matrix.
[0084] In one embodiment, the direction matrix can be used. Calculate the angular distance or Euclidean distance between any two point light sources and construct the distance matrix between the light sources. ,in express An identity matrix of order 1.
[0085] In the application process, the direction matrix P and its transpose P are calculated. T The product of these two matrices yields the matrix S=PP. T The elements S in matrix S ij =P i .P j The cosine similarity is represented by the dot product (cosine value) between the direction vectors of the i-th and j-th light sources, which represents the cosine similarity of the angle between the two light source directions. To more intuitively represent the "distance", the similarity matrix S is subtracted by an n-order identity matrix I. nThe distance matrix D=SI is obtained. n In the distance matrix D, the diagonal elements become 0 (distances between themselves), while the off-diagonal elements become 0. ij This represents the cosine of the angle between the directions of light source i and light source j. The larger the value (the closer to 1), the closer the directions of the two light sources are, meaning the spatial "distance" is closer; the smaller the value, the farther apart the directions of the two light sources are.
[0086] To evaluate the distance between each light source and its nearest neighbor, key information needs to be extracted from the distance matrix D. Specifically, this involves taking the largest element from each row of the distance matrix, i.e., along the distance matrix between light sources. The maximum value of a row or column is taken to obtain the distance vector representing the light source distribution density. For the i-th row of the distance matrix D, iterate through all elements in that row (j from 1 to n), find the maximum value, and denote it as d. i =maxD ij Because of D ii =0, the maximum value in this row must come from the comparison result of this light source with all other light sources. d i The physical meaning of is: the minimum cosine value (i.e., the value closest to 1) of the angle between the i-th light source and all other light sources; it directly reflects the local density at the location of that light source. The d values of all n light sources... i Arranged in order, they form the distance vector.
[0087] In another exemplary embodiment of the present invention, the following description continues using the previously described embodiments as examples. Based on the spherical Gaussian sampling kernel, the target dynamic range ambient light map is sampled to obtain the RGB channel values that match the target dynamic range ambient light map corresponding to the fixed light source. This can be achieved by sampling in the following manner:
[0088] Based on the spherical Gaussian sampling kernel, the target dynamic range ambient light map is sampled to obtain multiple channel values under each RGB channel;
[0089] For each RGB channel, the average value of multiple channel values under each channel is calculated to obtain the channel average value for each channel;
[0090] Based on the average value of each channel, the RGB channel values of the target dynamic range ambient light map corresponding to the fixed light source are obtained.
[0091] In one embodiment, for a target dynamic range ambient light map (img), this map represents the incident light intensity in each direction v of a sphere using a mapping method such as isometric cylindrical projection or cube mapping. The target dynamic range ambient light map (img) is discretely sampled using this spherical Gaussian sampling kernel. Specifically, a sufficiently dense and uniform set of discrete sampling directions {v1, v2, ..., vk} is selected or generated on the sphere. For each sampling direction vj, the following operations can be performed:
[0092] Find the pixel value corresponding to direction vj in the ambient light map img. This value usually contains three independent brightness components, namely the red channel value, the green channel value, and the blue channel value.
[0093] Calculate the weight of the sampling kernel in this direction;
[0094] The weights are applied to the three channels respectively to obtain the contribution of the sampling point to the three channels;
[0095] After traversing all k sampling directions, the average value is calculated for each RGB channel to obtain the channel average value for each channel.
[0096] The average values of the three channels are combined to form the RGB channel values of the target dynamic range ambient light map corresponding to the i-th fixed light source. In other words, the RGB channel values of the target dynamic range ambient light map corresponding to the fixed light source are obtained based on the average values of each channel.
[0097] In this embodiment, a spherical Gaussian sampling kernel is used to perform weighted sampling on the ambient light map. This method not only considers the direction of the light source but also intelligently incorporates the lighting information from its surrounding directions. By independently performing weighted averaging on each color channel, the average color of the light source in the map, which is most relevant to the region in the direction of the light source, can be accurately extracted. This allows a single point light source to best represent the ambient lighting characteristics of its corresponding direction and neighborhood, greatly improving the accuracy and realism of lighting reproduction.
[0098] Based on the same inventive concept, the present invention also provides an image acquisition method under dynamic ambient lighting. Figure 4 This is a flowchart illustrating the image acquisition method under dynamic ambient lighting provided by the present invention. The following will combine... Figure 4 The process of the image acquisition method under dynamic ambient lighting provided by the present invention will be described.
[0099] In an exemplary embodiment of the present invention, combined with Figure 4As can be seen, the image acquisition method under dynamic ambient lighting can be applied to image acquisition systems under dynamic ambient lighting. The image acquisition method under dynamic ambient lighting may include steps 410 to 430, which will be described in detail below.
[0100] In step 410, according to the ambient lighting encoding method, the target dynamic range ambient light map is encoded into a light source control signal with preset RGB channel values that matches the target dynamic range ambient light map.
[0101] In one embodiment, a target high dynamic range (HDR) ambient light map (img) serving as a lighting blueprint can be received. Subsequently, the map is processed according to the ambient light encoding method described in any of the preceding specific embodiments. The core of this encoding method is: using the center of the spherical architecture as the origin, and utilizing pre-calibrated light source direction information, intelligent sampling of the target map is performed using a spherical Gaussian function to calculate the precise color values (RGB channel values) emitted by each fixed light source on the sphere to reproduce the lighting effect of the map. Finally, the encoding method combines these color values with the hardware address of each light source to generate a light source control signal that can be directly executed by the hardware. This signal defines the preset RGB channel values for all light sources.
[0102] In step 420, based on the light source control signal, the fixed light source is controlled to switch the illumination state so that each of the fixed light sources independently provides a light source with preset RGB channel values for the object to be photographed.
[0103] In one embodiment, the control module sends the generated light source control signal to all fixed light sources on the spherical structure. Each light source receives the instruction based on its own address in the signal, and at a precise moment triggered by a synchronization control signal, simultaneously and independently switches its emission state to the RGB color and brightness specified by the instruction. After switching, all light sources collectively superimpose at the center of the sphere to form a preset ambient lighting condition that matches the target HDR ambient light map, providing illumination for the object to be photographed at the center of the sphere.
[0104] In step 430, when each of the fixed light sources independently provides a light source with a preset RGB channel value for the object to be photographed, the shooting module is triggered to photograph the object to be photographed, thereby obtaining an image of the object to be photographed under the preset ambient light.
[0105] In one embodiment, at the same moment the fixed light source completes its state switching and stably provides the preset ambient light, the control module triggers the shooting module via a synchronous control signal. The shooting module, such as one or more cameras, then captures an image of the object to be photographed located at the center of the sphere. Because the lighting and the shooting action are strictly synchronized, the image captured by the camera is the image of the object to be photographed under the preset ambient light.
[0106] In another embodiment, if it is necessary to acquire images of the same object under various different ambient lighting conditions, steps 410 to 430 can be repeated. That is, multiple HDR ambient light maps are encoded sequentially or in sequence into multiple sets of light source control signals. Then, the control module executes steps 420 to 430 sequentially at set time intervals (such as video frame periods), thereby continuously acquiring multiple object images under different preset lighting conditions and with pixel-level alignment, forming an image sequence or video.
[0107] In this embodiment, it is possible to reproduce in physical space, with high fidelity and repeatability, any environmental lighting that originally existed only in computers and was arbitrarily complex or even difficult to arrange in reality, such as outdoor skylight at a specific moment or special artistic lighting effects, and to complete the shooting of real objects, greatly expanding the range and realism of the lighting conditions that can be obtained.
[0108] In another exemplary embodiment of the present invention, the above-described embodiments will continue to be used as examples for explanation. The target dynamic range ambient light map may include multiple target dynamic range ambient light maps. The step of encoding the target dynamic range ambient light map into a light source control signal with preset RGB channel values that matches the target dynamic range ambient light map, according to any one of the ambient light encoding methods described above, can be implemented in the following manner:
[0109] According to any one of the ambient lighting encoding methods, multiple target dynamic range ambient light maps are encoded into a light source control signal sequence with preset RGB channel values that matches the multiple target dynamic range ambient light maps;
[0110] Based on the light source control signal, the fixed light sources are controlled to switch illumination states so that each fixed light source independently provides a light source with preset RGB channel values for the subject to be photographed. This can be achieved in the following way:
[0111] Based on the light source control signal sequence, the fixed light source is controlled to switch the illumination state so that each fixed light source independently provides a light source with preset RGB channel values for the object to be photographed in the order of the light source control signal sequence.
[0112] When each of the fixed light sources independently provides a light source with preset RGB channel values for the object to be photographed, triggering the shooting module to photograph the object to be photographed and obtaining an image of the object to be photographed under preset ambient lighting can be achieved in the following way:
[0113] When each of the fixed light sources independently provides a light source with a preset RGB channel value to the object to be photographed in the order of the light source control signal sequence, the shooting module is triggered to photograph the object to be photographed, and an image sequence of the object to be photographed under the preset ambient light is obtained.
[0114] A video clip is obtained by combining the image sequence of the object to be photographed under preset ambient lighting conditions.
[0115] In one embodiment, a sequence of m target dynamic range ambient light maps can be received, where each map represents a different preset ambient lighting. For each map in the sequence, the ambient lighting encoding method described in any of the preceding specific embodiments is processed. That is, sampling encoding based on a spherical Gaussian function is performed independently on each map to calculate the RGB channel values required by each fixed light source on the spherical architecture to reproduce the lighting of that map, and these values are converted into binary control instructions containing the light source address. After processing all m maps, m sets of sequentially arranged control instructions are obtained, which together constitute a light source control signal sequence. This sequence completely defines all the lighting states that the light source system needs to present over time (in frame order).
[0116] Furthermore, the control module no longer sends a single light source control signal, but instead sends each set of instructions in the light source control signal sequence sequentially. During application, the instruction corresponding to the first frame (t=1) in the sequence can be sent to all light sources and buffered. Triggered by the first synchronization pulse, all light sources switch to the state defined by the instruction in the first frame, providing the first preset ambient lighting for the subject. Subsequently, under the control of the system clock, the second, third, ... up to the m-th set of instructions in the sequence are sent and triggered sequentially at fixed, high-precision intervals. This allows all fixed light sources to periodically and synchronously switch their emission states according to the order of the light source control signal sequence, dynamically providing a series of different, independent preset RGB light source illuminations for the subject.
[0117] In another embodiment, at each synchronization moment when the light source switches states sequentially, the shooting module is synchronously triggered to capture the object to be photographed at the center of the sphere. If the m images represent n different lighting scenes, and each scene requires capturing multiple consecutive frames (e.g., k frames) to form a longer video, then the above light source control signal sequence can be regarded as a "lighting mode" loop of length m. By continuously looping the entire sequence k times while synchronously triggering the camera at a frame rate of f, a total of m×k images can be captured. Then, the images are reassembled according to the shooting time to obtain n complete video segments (each segment containing k frames), each video segment showing the continuous state of the object under a specific dynamic lighting condition.
[0118] In this embodiment, the physical light source system can be driven to reproduce a series of complex lighting changes automatically and according to a pre-programmed script, while simultaneously recording video of the object. This provides an efficient and controllable physical method for generating paired video data of the same dynamic object under different temporal lighting conditions on a large scale, solving the core problem of the near-absence of supervised video data in cutting-edge research areas such as neural relighting of dynamic scenes and dynamic material estimation.
[0119] Figure 5 This is a schematic diagram of the ambient light encoding device provided by the present invention.
[0120] The ambient light encoding device provided by the present invention is described below. The ambient light encoding device described below can be referred to in correspondence with the ambient light encoding method described above.
[0121] In an exemplary embodiment of the present invention, combined with Figure 5 As can be seen, the ambient light encoding device is applied to any of the dynamic ambient light image acquisition systems described above. The device includes a calibration module 510, a determination module 520, a setting module 530, a processing module 540, a combination module 550, and a generation module 560. Each module will be described in detail below.
[0122] The calibration module 510 can be configured to calibrate the positions of multiple fixed light sources in the spherical structure with the center of the sphere as the origin and perform normalization processing to obtain a direction matrix that matches the position of the fixed light source, wherein the elements in the direction matrix are used to characterize the light source direction of the fixed light source.
[0123] The determining module 520 can be configured to determine a distance vector based on the direction matrix, wherein the distance vector is used to characterize the light source distribution density of the fixed light source;
[0124] The setting module 530 can be configured to set the kurtosis parameter of the spherical Gaussian function based on the distance vector;
[0125] The processing module 540 can be configured to traverse all fixed light sources and perform the following steps for any fixed light source: determine a spherical Gaussian sampling kernel based on the light source direction and the steepness parameter of the fixed light source; sample the target dynamic range ambient light map based on the spherical Gaussian sampling kernel to obtain the RGB channel values of the target dynamic range ambient light map corresponding to the fixed light source, wherein the target dynamic range ambient light map is the light map that matches the image of the object to be photographed under a preset ambient light.
[0126] The combination module 550 can be configured to combine the RGB channel values corresponding to each of the fixed light sources to obtain a combined RGB channel value sequence.
[0127] The generation module 560 can be configured to perform encoding processing based on the combined RGB channel value sequence to obtain a binary code of a light source with preset RGB channel values that matches the target dynamic range ambient light map, and to obtain a light source control signal based on the binary code.
[0128] In an exemplary embodiment of the present invention, the determining module 520 may determine the distance vector based on the direction matrix in the following manner:
[0129] Based on the direction matrix, a distance matrix between different light sources is determined, wherein the elements in the distance matrix are used to characterize the distance between different fixed light sources;
[0130] Take the largest column element of each column along the distance matrix, and / or take the largest row element of each row along the distance matrix;
[0131] The distance vector is determined based on the largest column elements and / or the largest row elements of the matrix.
[0132] In an exemplary embodiment of the present invention, the processing module 540 may sample the target dynamic range ambient light map based on the spherical Gaussian sampling kernel to obtain RGB channel values that match the target dynamic range ambient light map corresponding to the fixed light source:
[0133] Based on the spherical Gaussian sampling kernel, the target dynamic range ambient light map is sampled to obtain multiple channel values under each RGB channel;
[0134] For each RGB channel, the average value of multiple channel values under each channel is calculated to obtain the channel average value for each channel;
[0135] Based on the average value of each channel, the RGB channel values of the target dynamic range ambient light map corresponding to the fixed light source are obtained.
[0136] Figure 6 This is a schematic diagram of the image acquisition device under dynamic ambient lighting provided by the present invention.
[0137] The image acquisition device under dynamic ambient lighting provided by the present invention is described below. The image acquisition device under dynamic ambient lighting described below can be referred to in correspondence with the image acquisition method under dynamic ambient lighting described above.
[0138] In an exemplary embodiment of the present invention, combined with Figure 6 As can be seen, the image acquisition device under dynamic ambient lighting is applied to any of the image acquisition systems under dynamic ambient lighting described in this article; the image acquisition device under dynamic ambient lighting may include an encoding module 610, a switching module 620, and an acquisition module 630, and each module will be described in detail below.
[0139] Encoding module 610 can be configured to encode a target dynamic range ambient light map into a light source control signal with preset RGB channel values that matches the target dynamic range ambient light map according to any of the ambient light encoding methods described above.
[0140] The switching module 620 can be configured to control the fixed light source to switch the illumination state based on the light source control signal, so that each fixed light source independently provides a light source with a preset RGB channel value for the object to be photographed.
[0141] The acquisition module 630 can be configured to trigger the shooting module to capture the object under preset ambient light when each of the fixed light sources independently provides a light source with preset RGB channel values for the object to be photographed.
[0142] In an exemplary embodiment of the present invention, the target dynamic range ambient light map includes multiple target dynamic range ambient light maps; the encoding module 610 can encode the target dynamic range ambient light map into a light source control signal with preset RGB channel values that matches the target dynamic range ambient light map using any of the ambient light encoding methods described in the present invention:
[0143] According to any one of the ambient lighting encoding methods, multiple target dynamic range ambient light maps are encoded into a light source control signal sequence with preset RGB channel values that matches the multiple target dynamic range ambient light maps;
[0144] The switching module 620 can control the fixed light source to switch its illumination state based on the light source control signal in the following way, so that each fixed light source independently provides a light source with preset RGB channel values for the object to be photographed:
[0145] Based on the light source control signal sequence, the fixed light source is controlled to switch the illumination state so that each fixed light source independently provides a light source with preset RGB channel values for the object to be photographed in the order of the light source control signal sequence.
[0146] The acquisition module 630 can achieve the following: when each of the fixed light sources independently provides a light source with preset RGB channel values for the object to be photographed, the shooting module is triggered to photograph the object to be photographed, thereby obtaining an image of the object to be photographed under preset ambient lighting conditions:
[0147] When each of the fixed light sources independently provides a light source with a preset RGB channel value to the object to be photographed in the order of the light source control signal sequence, the shooting module is triggered to photograph the object to be photographed, and an image sequence of the object to be photographed under the preset ambient light is obtained.
[0148] A video clip is obtained by combining the image sequence of the object to be photographed under preset ambient lighting conditions.
[0149] Figure 7 This is a schematic diagram of the electronic device provided by the present invention, such as... Figure 7As shown, the electronic device may include: a processor 710, a communications interface 720, a memory 730, and a communications bus 740, wherein the processor 710, the communications interface 720, and the memory 730 communicate with each other through the communications bus 740. Processor 710 can call logic instructions in memory 730 to execute an ambient lighting encoding method. This method is applied to any of the dynamic ambient lighting image acquisition systems described above. The method includes: using the center of a spherical structure as the origin, calibrating the positions of multiple fixed light sources within the spherical structure and performing normalization processing to obtain a direction matrix matching the positions of the fixed light sources, wherein the elements in the direction matrix characterize the light source direction of the fixed light source; determining a distance vector based on the direction matrix, wherein the distance vector characterizes the light source distribution density of the fixed light source; setting a kurtosis parameter of a spherical Gaussian function based on the distance vector; traversing all fixed light sources and performing the following steps for any fixed light source: determining a spherical Gaussian sampling kernel based on the light source direction of the fixed light source and the kurtosis parameter; sampling a target dynamic range ambient light map based on the spherical Gaussian sampling kernel to obtain RGB channel values matching the target dynamic range ambient light map corresponding to the fixed light source, wherein the target dynamic range ambient light map is obtained by capturing the object under a preset ambient light. The system captures a light map matching the image; combines the RGB channel values corresponding to each of the fixed light sources to obtain a combined RGB channel value sequence; encodes the combined RGB channel value sequence to obtain a binary code for a light source with preset RGB channel values that matches the target dynamic range ambient light map, and obtains a light source control signal based on the binary code; or, it executes a dynamic ambient light image acquisition method, which is applied to any of the dynamic ambient light image acquisition systems described above, the method comprising: encoding the target dynamic range ambient light map into a light source control signal for a light source with preset RGB channel values that matches the target dynamic range ambient light map according to an ambient light encoding method; controlling the fixed light sources to switch illumination states based on the light source control signal, so that each of the fixed light sources independently provides a light source with preset RGB channel values for the object to be photographed; triggering the shooting module to photograph the object to be photographed when each of the fixed light sources independently provides a light source with preset RGB channel values for the object to be photographed, thereby obtaining an image of the object to be photographed under preset ambient light.
[0150] Furthermore, the logical instructions in the aforementioned memory 730 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0151] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the ambient lighting encoding method provided by the above methods. The method is applied to any of the dynamic ambient lighting image acquisition systems described above. The method includes: using the center of the sphere of the spherical structure as the origin, calibrating the positions of multiple fixed light sources in the spherical structure and performing normalization processing to obtain a direction matrix matching the positions of the fixed light sources, wherein the elements in the direction matrix are used to characterize the light source direction of the fixed light source; determining a distance vector according to the direction matrix, wherein the distance vector is used to characterize the light source distribution density of the fixed light source; setting the kurtosis parameter of the spherical Gaussian function based on the distance vector; traversing all fixed light sources, and performing the following steps for any fixed light source: determining a spherical Gaussian sampling kernel based on the light source direction of the fixed light source and the kurtosis parameter; sampling the target dynamic range ambient light map based on the spherical Gaussian sampling kernel to obtain the RGB channel values corresponding to the fixed light source that match the target dynamic range ambient light map. In this process, the target dynamic range ambient light map is a light map that matches the image of the object to be photographed under a preset ambient light condition; the RGB channel values corresponding to each of the fixed light sources are combined to obtain a combined RGB channel value sequence; the combined RGB channel value sequence is encoded to obtain a binary code of a light source with preset RGB channel values that matches the target dynamic range ambient light map, and a light source control signal is obtained based on the binary code; or, a dynamic ambient light condition image acquisition method is executed, the method being applied to any one of the dynamic ambient light condition image acquisition systems described above, the method comprising: encoding the target dynamic range ambient light map into a light source control signal of a light source with preset RGB channel values that matches the target dynamic range ambient light map according to an ambient light encoding method; controlling the fixed light sources to switch illumination states based on the light source control signal, so that each of the fixed light sources independently provides a light source with preset RGB channel values for the object to be photographed; when each of the fixed light sources independently provides a light source with preset RGB channel values for the object to be photographed, triggering the shooting module to photograph the object to be photographed, thereby obtaining an image of the object to be photographed under the preset ambient light condition.
[0152] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the ambient lighting encoding method provided by the methods described above. The method is applied to any one of the dynamic ambient lighting image acquisition systems described above. The method includes: using the center of a spherical structure as the origin, calibrating the positions of multiple fixed light sources within the spherical structure and performing normalization processing to obtain a direction matrix matching the positions of the fixed light sources, wherein the elements in the direction matrix characterize the light source direction of the fixed light source; determining a distance vector based on the direction matrix, wherein the distance vector characterizes the light source distribution density of the fixed light source; setting a kurtosis parameter of a spherical Gaussian function based on the distance vector; traversing all fixed light sources, and for any fixed light source, performing the following steps: determining a spherical Gaussian sampling kernel based on the light source direction of the fixed light source and the kurtosis parameter; sampling a target dynamic range ambient light map based on the spherical Gaussian sampling kernel to obtain RGB channel values corresponding to the fixed light source that match the target dynamic range ambient light map, wherein the target dynamic range ambient light map... A light map matching the image under preset ambient lighting is obtained by capturing an image of the subject to be photographed; the RGB channel values corresponding to each of the fixed light sources are combined to obtain a combined RGB channel value sequence; the combined RGB channel value sequence is encoded to obtain a binary code of a light source with preset RGB channel values that matches the target dynamic range ambient light map, and a light source control signal is obtained based on the binary code; or, a dynamic ambient lighting image acquisition method is executed, the method being applied to any one of the dynamic ambient lighting image acquisition systems described above, the method comprising: encoding the target dynamic range ambient light map into a light source control signal with preset RGB channel values that matches the target dynamic range ambient light map according to an ambient lighting encoding method; controlling the fixed light sources to switch lighting states based on the light source control signal, so that each of the fixed light sources independently provides a light source with preset RGB channel values for the subject to be photographed; when each of the fixed light sources independently provides a light source with preset RGB channel values for the subject to be photographed, triggering the shooting module to capture the subject to be photographed, thereby obtaining an image of the subject to be photographed under preset ambient lighting.
[0153] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.
[0154] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.
[0155] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. An ambient lighting encoding method, characterized in that, The method includes: Using the center of the sphere of the spherical structure as the origin, the positions of multiple fixed light sources in the spherical structure are calibrated and normalized to obtain a direction matrix that matches the position of the fixed light source. The elements in the direction matrix are used to characterize the light source direction of the fixed light source. Based on the direction matrix, a distance vector is determined, wherein the distance vector is used to characterize the light source distribution density of the fixed light source; Based on the distance vector, the kurtosis parameter of the spherical Gaussian function is set; Iterate through all fixed light sources, and for any fixed light source, perform the following steps: Based on the light source direction of the fixed light source and the steepness parameter, a spherical Gaussian sampling kernel is determined; based on the spherical Gaussian sampling kernel, the target dynamic range ambient light map is sampled to obtain the RGB channel values that match the target dynamic range ambient light map corresponding to the fixed light source, wherein the target dynamic range ambient light map is the light map that matches the image of the object to be photographed under preset ambient lighting conditions. The RGB channel values corresponding to each of the fixed light sources are combined to obtain a combined RGB channel value sequence. Encoding is performed based on the combined RGB channel value sequence to obtain a binary code of a light source with preset RGB channel values that matches the target dynamic range ambient light map, and a light source control signal is obtained based on the binary code.
2. The ambient lighting encoding method according to claim 1, characterized in that, Determining the distance vector based on the direction matrix includes: Based on the direction matrix, a distance matrix between different light sources is determined, wherein the elements in the distance matrix are used to characterize the distance between different fixed light sources; Take the largest column element of each column along the distance matrix, and / or take the largest row element of each row along the distance matrix; The distance vector is determined based on the largest column elements and / or the largest row elements of the matrix.
3. The ambient lighting encoding method according to claim 1, characterized in that, The step of sampling the target dynamic range ambient light map based on the spherical Gaussian sampling kernel to obtain the RGB channel values of the target dynamic range ambient light map that match the fixed light source includes: Based on the spherical Gaussian sampling kernel, the target dynamic range ambient light map is sampled to obtain multiple channel values under each RGB channel; For each RGB channel, the average value of multiple channel values under each channel is calculated to obtain the channel average value for each channel; Based on the average value of each channel, the RGB channel values of the target dynamic range ambient light map corresponding to the fixed light source are obtained.
4. A method for image acquisition under dynamic ambient lighting, characterized in that, The method includes: According to any one of claims 1-3, the ambient lighting encoding method encodes the target dynamic range ambient light map into a light source control signal with a preset RGB channel value that matches the target dynamic range ambient light map. Based on the light source control signal, the fixed light source is controlled to switch the illumination state so that each fixed light source independently provides a light source with preset RGB channel values for the object to be photographed. When each of the fixed light sources independently provides a light source with preset RGB channel values for the object to be photographed, the shooting module is triggered to photograph the object to be photographed, and an image of the object to be photographed under preset ambient light is obtained.
5. The image acquisition method under dynamic ambient lighting according to claim 4, characterized in that, The target dynamic range ambient light map includes multiple target dynamic range ambient light maps; the ambient light encoding method according to any one of claims 1-3 encodes the target dynamic range ambient light map into a light source control signal with preset RGB channel values that matches the target dynamic range ambient light map, including: According to any one of claims 1-3, the ambient lighting encoding method encodes multiple target dynamic range ambient light maps into a light source control signal sequence with preset RGB channel values that matches the multiple target dynamic range ambient light maps. Based on the light source control signal, the fixed light sources are controlled to switch illumination states so that each fixed light source independently provides a light source with preset RGB channel values for the subject to be photographed, including: Based on the light source control signal sequence, the fixed light source is controlled to switch the illumination state so that each fixed light source independently provides a light source with preset RGB channel values for the object to be photographed in the order of the light source control signal sequence. When each of the fixed light sources independently provides a light source with preset RGB channel values for the object to be photographed, triggering the shooting module to photograph the object to be photographed, and obtaining an image of the object to be photographed under preset ambient lighting, includes: When each of the fixed light sources independently provides a light source with a preset RGB channel value to the object to be photographed in the order of the light source control signal sequence, the shooting module is triggered to photograph the object to be photographed, and an image sequence of the object to be photographed under the preset ambient light is obtained. A video clip is obtained by combining the image sequence of the object to be photographed under preset ambient lighting conditions.
6. An image acquisition system under dynamic ambient lighting, characterized in that, The system includes: A spherical structure, wherein the object to be photographed is placed at the center of the sphere of the spherical structure; Multiple fixed light sources are evenly distributed on the spherical surface of the spherical structure, and each fixed light source independently provides a light source with preset RGB channel values for the object to be photographed; A control module is used to send a light source control signal to the fixed light source, so that the fixed light source independently provides a light source with a preset RGB channel value for the object to be photographed after receiving the light source control signal, wherein the light source control signal is implemented using the ambient light encoding method according to any one of claims 1-3; The shooting module, whose shooting direction is pointed to the center of the sphere of the spherical structure, is used to shoot the subject under the condition that each of the fixed light sources independently provides light sources with preset RGB channel values to the subject, so as to obtain an image of the subject under preset ambient light, wherein the preset ambient light is formed by each of the fixed light sources independently providing light sources with preset RGB channel values to the subject.
7. The image acquisition system under dynamic ambient lighting according to claim 6, characterized in that, The control module also includes: The synchronization control module is used to send synchronization control signals to the fixed light source and the imaging module so that the switching of the fixed light source and the image acquisition of the imaging module are performed synchronously.
8. The image acquisition system under dynamic ambient lighting according to claim 6 or 7, characterized in that, The spherical structure includes a regular 32-sided polyhedron, which includes multiple vertices and multiple midpoints of edges; the fixed light source is respectively set at the vertex and / or the midpoint of the edge.
9. The image acquisition system under dynamic ambient lighting according to claim 7, characterized in that, The fixed light source is composed of an LED bead matrix; the LED bead matrix is controlled by binary encoding; the LED bead matrix has preset RGB channel values.
10. The image acquisition system under dynamic ambient lighting according to claim 6, characterized in that, The fixed light source is also configured with a light source address; the control module sends the light source control signal to the fixed light source through the light source address.
11. An ambient light encoding device, characterized in that, The device includes: The calibration module is used to calibrate the positions of multiple fixed light sources in the spherical structure with the center of the sphere as the origin and perform normalization processing to obtain a direction matrix that matches the position of the fixed light source. The elements in the direction matrix are used to characterize the light source direction of the fixed light source. The determining module is used to determine a distance vector based on the direction matrix, wherein the distance vector is used to characterize the light source distribution density of the fixed light source; The setting module is used to set the kurtosis parameter of the spherical Gaussian function based on the distance vector; The processing module is used to traverse all fixed light sources and perform the following steps for any fixed light source: determine a spherical Gaussian sampling kernel based on the light source direction and the steepness parameter of the fixed light source; sample the target dynamic range ambient light map based on the spherical Gaussian sampling kernel to obtain the RGB channel values of the target dynamic range ambient light map that match the fixed light source, wherein the target dynamic range ambient light map is the light map that matches the image of the object to be photographed under a preset ambient light. The combination module is used to combine the RGB channel values corresponding to each of the fixed light sources to obtain a combined RGB channel value sequence. The generation module is used to perform encoding processing based on the combined RGB channel value sequence to obtain the binary code of the light source with preset RGB channel values that matches the target dynamic range ambient light map, and to obtain the light source control signal based on the binary code.
12. An image acquisition device under dynamic ambient lighting, characterized in that, The device includes: The encoding module is used to encode the target dynamic range ambient light map into a light source control signal with preset RGB channel values that matches the target dynamic range ambient light map according to the ambient light encoding method of any one of claims 1-3. The switching module is used to control the fixed light source to switch the illumination state based on the light source control signal, so that each fixed light source independently provides a light source with preset RGB channel values for the object to be photographed; The acquisition module is used to trigger the shooting module to capture the object under preset ambient light when each of the fixed light sources independently provides a light source with preset RGB channel values for the object to be photographed.
13. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to execute the ambient lighting encoding method according to any one of claims 1 to 3, or the image acquisition method under dynamic ambient lighting according to any one of claims 4 to 5, through the computer program.