CN115438413B - Light source arrangement method, electronic device, and computer-readable storage medium - Google Patents
Light source arrangement method, electronic device, and computer-readable storage medium Download PDFInfo
- Publication number
- CN115438413B CN115438413B CN202211123873.2A CN202211123873A CN115438413B CN 115438413 B CN115438413 B CN 115438413B CN 202211123873 A CN202211123873 A CN 202211123873A CN 115438413 B CN115438413 B CN 115438413B
- Authority
- CN
- China
- Prior art keywords
- light source
- sample
- space
- information
- subspace
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Abstract
The embodiment of the disclosure discloses a light source layout method, electronic equipment and a computer readable storage medium, wherein the method comprises the following steps: obtaining space information data of a target space; wherein the target space comprises at least one subspace; determining sub-information data for each of the sub-spaces based on the spatial information data; obtaining a plan view of each subspace based on the subspace information data of each subspace; processing each plane map based on a preset network model respectively, and determining a light source distribution map of each subspace; according to the embodiment, the plane diagram corresponding to the subspace is processed through the deep neural network, so that the light source distribution diagram with at least one light source arranged in the subspace can be directly obtained, automatic arrangement of the light source in the space is realized, the method is more suitable for industrial production, and the consumption of manpower and material resources caused by artificial arrangement of the light source in the prior art is overcome.
Description
Technical Field
The present disclosure relates to a light source arrangement method, an electronic device, and a computer-readable storage medium.
Background
In indoor design scenes, the design software is usually provided with tools for distributing and adjusting indoor light, and for different house types and indoor furniture placement scenes, the designer is usually required to manually add and modify the light types and parameters, so that a better rendering effect is finally achieved; however, the manual arrangement of the light source has high requirements for designers and is time-consuming and labor-consuming.
Disclosure of Invention
According to an aspect of the embodiments of the present disclosure, there is provided a light source arrangement method including:
obtaining space information data of a target space; wherein the target space comprises at least one subspace;
determining sub-information data for each of the sub-spaces based on the spatial information data;
obtaining a plan view of each subspace based on the subspace information data of each subspace;
processing each plane map based on a preset network model respectively, and determining a light source distribution map of each subspace; wherein the light source distribution map comprises at least one light source.
Optionally, the determining sub information data of each of the subspaces based on the spatial information data includes:
carrying out structuring treatment on the space information data to obtain structured space data;
And decomposing the structured space data to determine sub-information data of each sub-space.
Optionally, the obtaining a plan view of each subspace based on the subspace information data of each subspace includes:
for each subspace, acquiring top view information corresponding to the subspace based on the structured subspace information data;
extracting space edge information of the subspace and equipment information of at least one piece of equipment in the subspace based on the top view information;
and obtaining a plan view of the subspace based on the space edge information and the at least one piece of equipment information.
Optionally, before processing each of the plan views based on a preset network model, determining a distribution diagram of the light source of each subspace, the method further includes:
training the preset network model based on a sample space set; wherein the set of sample spaces comprises at least one sample space.
Optionally, the training the preset network model based on the sample space set includes:
preprocessing at least one sample space in the sample space set to obtain a sample image set and a supervision image set corresponding to the sample image set; the sample image set comprises at least one sample image, the supervision image set comprises at least one supervision image, and the sample images are in one-to-one correspondence with the supervision images;
Processing the sample image based on the preset network model to obtain a predicted light source distribution diagram;
and training the preset network model based on the predicted light source distribution diagram and the network loss determined by the supervision image.
Optionally, the preprocessing at least one sample space in the sample space set to obtain a sample image set and a supervision image set corresponding to the sample image set includes:
obtaining a sample space data set based on sample space data of each sample space in the sample space set; wherein the sample space data comprises space edge information in the sample space, at least one piece of equipment information in the sample space and at least one piece of light source information in the sample space;
and carrying out imaging processing on each sample space data in the sample space data set to obtain the sample image set and the supervision image set.
Optionally, the performing an imaging process on each sample space data in the sample space data set to obtain the sample image set and the supervision image set includes:
carrying out structuring treatment on each sample space data in the sample space data set to obtain structured sample space data;
Obtaining sample top view information corresponding to the sample space based on the structured sample space data;
extracting the spatial edge information, the at least one piece of equipment information and the at least one piece of light source information based on the sample top view information;
obtaining a sample image corresponding to the sample space based on the space edge information and the at least one piece of equipment information;
and obtaining a supervision image corresponding to the sample space based on the space edge information, the at least one piece of equipment information and the at least one piece of light source information.
Optionally, the training the preset network model based on the predicted light source distribution diagram and the network loss determined by the monitoring image includes:
determining the network loss based on at least one predicted light source detection box included in the predicted light source profile and at least one supervising light source detection box included in the supervising image; wherein the predicted light source detection frames correspond to the supervision light source detection frames one by one;
training the preset network model based on the network loss.
Optionally, the determining the network loss based on at least one predicted light source detection box included in the predicted light source distribution map and at least one supervised light source detection box included in the supervised image includes:
Determining a first loss based on an intersection ratio between each of the at least one predicted light source detection frames and the corresponding supervising light source detection frame;
determining a brightness difference between each predicted light source detection frame and the corresponding supervision light source detection frame in a color space dimension, and determining a second loss;
the network loss is determined based on the first loss and the second loss.
According to another aspect of the embodiments of the present disclosure, there is provided a light source arrangement apparatus including:
the data acquisition module is used for acquiring space information data of the target space; wherein the target space comprises at least one subspace;
a sub-information module for determining sub-information data of each of the sub-spaces based on the spatial information data; the plan view module is used for obtaining a plan view of each subspace based on the subspace information data of each subspace;
the light source distribution module is used for respectively processing each plane graph based on a preset network model and determining a light source distribution diagram of each subspace; wherein the light source distribution map comprises at least one light source.
Optionally, the sub information module includes:
The structuring unit is used for carrying out structuring processing on the space information data to obtain structured space data;
and the data decomposition unit is used for carrying out decomposition processing on the structured space data and determining the sub-information data of each subspace.
Optionally, the plan view module is specifically configured to obtain, for each subspace, plan view information corresponding to the subspace based on the structured sub-information data; extracting space edge information of the subspace and equipment information of at least one piece of equipment in the subspace based on the top view information; and obtaining a plan view of the subspace based on the space edge information and the at least one piece of equipment information.
Optionally, the apparatus further comprises:
the model training module is used for training the preset network model based on the sample space set; wherein the set of sample spaces comprises at least one sample space.
Optionally, the model training module includes:
the preprocessing unit is used for preprocessing at least one sample space in the sample space set to obtain a sample image set and a supervision image set corresponding to the sample image set; the sample image set comprises at least one sample image, the supervision image set comprises at least one supervision image, and the sample images are in one-to-one correspondence with the supervision images;
The light source prediction unit is used for processing the sample image based on the preset network model to obtain a predicted light source distribution diagram;
and the loss training unit is used for training the preset network model based on the predicted light source distribution diagram and the network loss determined by the supervision image.
Optionally, the preprocessing unit is specifically configured to obtain a sample space data set based on sample space data of each sample space in the sample space set; wherein the sample space data comprises space edge information in the sample space, at least one piece of equipment information in the sample space and at least one piece of light source information in the sample space; and carrying out imaging processing on each sample space data in the sample space data set to obtain the sample image set and the supervision image set.
Optionally, the preprocessing unit is configured to perform, when performing imaging processing on each sample space data in the sample space data set to obtain the sample image set and the supervisory image set, structuring processing on each sample space data in the sample space data set to obtain structured sample space data; obtaining sample top view information corresponding to the sample space based on the structured sample space data; extracting the spatial edge information, the at least one piece of equipment information and the at least one piece of light source information based on the sample top view information; obtaining a sample image corresponding to the sample space based on the space edge information and the at least one piece of equipment information; and obtaining a supervision image corresponding to the sample space based on the space edge information, the at least one piece of equipment information and the at least one piece of light source information.
Optionally, the loss training unit is specifically configured to determine the network loss based on at least one predicted light source detection frame included in the predicted light source distribution map and at least one supervised light source detection frame included in the supervised image; wherein the predicted light source detection frames correspond to the supervision light source detection frames one by one; training the preset network model based on the network loss.
Optionally, the loss training unit is configured to determine, when determining the network loss based on at least one predicted light source detection box included in the predicted light source distribution map and at least one supervised light source detection box included in the supervised image, a first loss based on an intersection ratio between each of the at least one predicted light source detection boxes and the corresponding supervised light source detection box; determining a brightness difference between each predicted light source detection frame and the corresponding supervision light source detection frame in a color space dimension, and determining a second loss; the network loss is determined based on the first loss and the second loss.
According to still another aspect of the embodiments of the present disclosure, there is provided an electronic device including:
A memory for storing a computer program product;
and a processor, configured to execute the computer program product stored in the memory, and when the computer program product is executed, implement the light source arrangement method according to any one of the foregoing embodiments.
According to a further aspect of the embodiments of the present disclosure, there is provided a computer-readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the light source arrangement method according to any of the embodiments described above.
According to a further aspect of the disclosed embodiments, there is provided a computer program product comprising computer program instructions which, when executed by a processor, implement the light source arrangement method according to any of the embodiments described above.
The technical scheme of the present disclosure is described in further detail below through the accompanying drawings and examples.
Drawings
The above and other objects, features and advantages of the present disclosure will become more apparent by describing embodiments thereof in more detail with reference to the accompanying drawings. The accompanying drawings are included to provide a further understanding of embodiments of the disclosure, and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure, without limitation to the disclosure. In the drawings, like reference numerals generally refer to like parts or steps.
FIG. 1 is a flow chart of a light source layout method provided by an exemplary embodiment of the present disclosure;
FIG. 2 is a schematic flow chart of step 104 in the embodiment of FIG. 1 of the present disclosure;
FIG. 3a is a schematic flow chart of step 106 in the embodiment of FIG. 1 of the present disclosure;
FIG. 3b is a plan view of a subspace provided by an exemplary embodiment of the present disclosure;
FIG. 3c is a light source distribution diagram obtained by light source layout according to the plan view provided in FIG. 3 b;
FIG. 4 is a flow chart of a light source layout method provided by another exemplary embodiment of the present disclosure;
FIG. 5 is a flow chart illustrating step 408 in the embodiment of FIG. 4 of the present disclosure;
FIG. 6 is a schematic flow chart of step 4081 in the embodiment of FIG. 5 of the present disclosure;
FIG. 7 is a schematic diagram of an apparatus for providing a light source arrangement according to an exemplary embodiment of the present disclosure;
fig. 8 is a block diagram of an electronic device provided in an exemplary embodiment of the present disclosure.
Detailed Description
Hereinafter, example embodiments according to the present disclosure will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present disclosure and not all of the embodiments of the present disclosure, and that the present disclosure is not limited by the example embodiments described herein.
It should be noted that: the relative arrangement of the components and steps, numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present disclosure unless it is specifically stated otherwise.
It will be appreciated by those of skill in the art that the terms "first," "second," etc. in embodiments of the present disclosure are used merely to distinguish between different steps, devices or modules, etc., and do not represent any particular technical meaning nor necessarily logical order between them.
It should also be understood that in embodiments of the present disclosure, "plurality" may refer to two or more, and "at least one" may refer to one, two or more.
It should also be appreciated that any component, data, or structure referred to in the presently disclosed embodiments may be generally understood as one or more without explicit limitation or the contrary in the context.
In addition, the term "and/or" in this disclosure is merely an association relationship describing an association object, and indicates that three relationships may exist, for example, a and/or B may indicate: a exists alone, A and B exist together, and B exists alone. In addition, the character "/" in the present disclosure generally indicates that the front and rear association objects are an or relationship. The data referred to in this disclosure may include unstructured data, such as text, images, video, and the like, as well as structured data.
It should also be understood that the description of the various embodiments of the present disclosure emphasizes the differences between the various embodiments, and that the same or similar features may be referred to each other, and for brevity, will not be described in detail.
Meanwhile, it should be understood that the sizes of the respective parts shown in the drawings are not drawn in actual scale for convenience of description.
The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the disclosure, its application, or uses.
Techniques, methods, and apparatus known to one of ordinary skill in the relevant art may not be discussed in detail, but are intended to be part of the specification where appropriate.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further discussion thereof is necessary in subsequent figures.
Embodiments of the present disclosure may be applicable to electronic devices such as terminal devices, computer systems, servers, etc., which may operate with numerous other general purpose or special purpose computing system environments or configurations. Examples of well known terminal devices, computing systems, environments, and/or configurations that may be suitable for use with the terminal device, computer system, server, or other electronic device include, but are not limited to: personal computer systems, server computer systems, thin clients, thick clients, hand-held or laptop devices, microprocessor-based systems, set-top boxes, programmable consumer electronics, network personal computers, minicomputer systems, mainframe computer systems, and distributed cloud computing technology environments that include any of the above systems, and the like.
Electronic devices such as terminal devices, computer systems, servers, etc. may be described in the general context of computer system-executable instructions, such as program modules, being executed by a computer system. Generally, program modules may include routines, programs, objects, components, logic, data structures, etc., that perform particular tasks or implement particular abstract data types. The computer system/server may be implemented in a distributed cloud computing environment in which tasks are performed by remote processing devices that are linked through a communications network. In a distributed cloud computing environment, program modules may be located in both local and remote computing system storage media including memory storage devices.
Exemplary method
Fig. 1 is a flowchart illustrating a light source arrangement method according to an exemplary embodiment of the present disclosure. The embodiment can be applied to an electronic device, as shown in fig. 1, and includes the following steps:
step 102, obtaining space information data of the target space.
Wherein the target space comprises at least one subspace.
Alternatively, the target space may be any space with a boundary where a light source needs to be arranged, where the boundary may be a solid boundary (e.g., a wall surface, a fence, a tree, etc.), or an edge of a space defined by man (e.g., a certain area where planning is performed by scribing, etc. as a target space); when the target space comprises a plurality of parts which can be connected through a small communication port relative to the target space, each part can be used as one subspace in the target space, for example, when the target space is a house, each room is communicated through a door, at the moment, for the light sources among different rooms to be mutually independent, each room can be used as one subspace of the house, the rationality of light source arrangement can be improved by arranging the light sources in each subspace, and the situation that certain subspaces are free of light sources due to the fact that light of the light sources cannot penetrate due to the interval among the subspaces is avoided.
Step 104, determining sub-information data of each sub-space based on the spatial information data.
Optionally, when the spatial information data is stored, the sub-information data for different sub-spaces may be stored separately, or based on the position information in the spatial information data, which information belongs to which sub-space may be determined, so as to obtain sub-information data corresponding to each sub-space, where the information related to the light source arrangement of each sub-space is described by the sub-information data, and the information may include, but is not limited to: location, size, internally located device location, etc.
And 106, obtaining a plan view of each subspace based on the subspace information data of each subspace.
In the embodiment, in order to realize the utilization of the deep neural network, each piece of sub-information data is converted into a plan view, and a preset network model is input in the form of an image to realize the prediction of light source arrangement; the acquisition of the plan view can be obtained by any drawing software of the prior art based on the sub information data.
And step 108, processing each plan view based on a preset network model respectively, and determining a light source distribution diagram of each subspace.
Wherein the light source distribution map comprises at least one light source.
The light source in this embodiment may be in any light source form, for example, a point light source, a surface light source, or the like, and in the light source distribution diagram, the light source may be represented by a rectangular frame with known position, size, and brightness information (for example, brightness may be represented by different gray scales), and the light source distribution diagram is identical to the shape of the plan view, except that at least one predicted light source is also included in the light source distribution diagram.
According to the light source layout method provided by the embodiment of the disclosure, spatial information data of a target space is obtained; wherein the target space comprises at least one subspace; determining sub-information data for each of the sub-spaces based on the spatial information data; obtaining a plan view of each subspace based on the subspace information data of each subspace; processing each plane map based on a preset network model respectively, and determining a light source distribution map of each subspace; wherein the light source distribution map comprises at least one light source; according to the embodiment, the plane diagram corresponding to the subspace is processed through the deep neural network, so that the light source distribution diagram with at least one light source arranged in the subspace can be directly obtained, automatic arrangement of the light source in the space is realized, the method is more suitable for industrial production, and the consumption of manpower and material resources caused by artificial arrangement of the light source in the prior art is overcome.
As shown in fig. 2, step 104 may include the following steps, based on the embodiment shown in fig. 1, described above:
step 1041, performing a structuring process on the spatial information data to obtain structured spatial data.
Optionally, structuring the spatial information data into json data to obtain structured spatial data; the spatial information data may be 3D information, including not only plane information but also height information.
In step 1042, the structured space data is decomposed to determine sub-information data of each sub-space.
Optionally, the sub-information data corresponding to each subspace is split from the structured space data through the position information in the structured space data to obtain at least one piece of sub-information data, the sub-information data comprises information such as the structure of the subspace, the size and the position of the internal equipment and the like, drawing of a plan view of the subspace can be achieved based on the sub-information data, and the drawing of the obtained plan view comprises the outline position of the subspace and the outline and the position of each equipment included in the plan view so as to facilitate light source arrangement of the subspace.
As shown in fig. 3a, based on the embodiment shown in fig. 1, step 106 may include the following steps for each subspace:
In step 1061, top view information corresponding to the subspace is obtained based on the structured sub-information data.
Optionally, the sub-information data is three-dimensional space information, and plan view information corresponding to the space information is obtained by extracting the plan view.
Step 1062, extracting spatial edge information of the subspace and device information of at least one device in the subspace based on the top view information.
The spatial edge information in the top view information may form a frame corresponding to the subspace, and the frame and the position of each device in the subspace may be determined based on at least one piece of device information, alternatively, in an optional example, when the subspace is a room, the device may be furniture, a large household appliance, and the like, and the spatial edge information may include wall line information, door information, window information, and the like.
At step 1063, a plan view of the subspace is obtained based on the spatial edge information and the at least one device information.
In this embodiment, in order to obtain a plan view, linearization may be performed on edge information having a certain width, so that the obtained plan view is a linearized plan view, for example, a plan view is drawn on a wall line at an inner wall position thereof, device information is abstracted into a regular-shaped plan frame, for example, similar to a house type drawing abstraction, a peripheral frame boundingbox of a device is abstracted into a plan view in a top view, for example, as shown in fig. 3b, by abstracting a top view corresponding to one subspace (room), a plan view represented by a line is obtained, wherein a rectangular frame represents a device in the subspace, for example, a door, a window, furniture, and the like; inputting the plan view shown in fig. 3b into a preset network model to obtain a light source distribution diagram shown in fig. 3c, wherein light sources with different intensities can be represented by different colors (or different brightnesses); in addition, lines of different nature may be represented in plan view in different colors, for example, lines of different colors and rectangular boxes are used to characterize wall lines, doors, windows, and the like.
In this embodiment, before executing step 1061, data cleaning may be further performed, and the structured sub-information data obtained by the structured processing may be determined by an automation script (according to a preset rule) whether to process standards of a house type structure, furniture, etc., and the sub-information data that does not conform to the preset rule is deleted, where the preset rule may include, but is not limited to: unreasonable house type (for example, only a wall of a room has no door or window, etc.), unreasonable furniture position, etc.
Fig. 4 is a flowchart illustrating a light source arrangement method according to another exemplary embodiment of the present disclosure. As shown in fig. 4, the method provided in this embodiment includes:
step 402, spatial information data of a target space is obtained.
Wherein the target space comprises at least one subspace.
The implementation and effect of this step can be understood with reference to step 102 in the embodiment shown in fig. 1, and will not be described herein.
Step 404, determining sub-information data for each sub-space based on the spatial information data.
The implementation and effect of this step can be understood with reference to step 104 in the embodiment shown in fig. 1, and will not be described herein.
Step 406, obtaining a plan view of each subspace based on the subspace information data of each subspace.
The implementation and effect of this step can be understood with reference to step 106 in the embodiment shown in fig. 1, and will not be described herein.
Step 408, training a preset network model based on the sample space set.
Wherein the set of sample spaces comprises at least one sample space, optionally, the distribution information of the light sources is known in each sample space, the sample spaces in the present embodiment correspond to the subspaces in the above embodiments, and each sample space does not need to be subjected to space division.
Step 410, processing each plan view based on the preset network model, and determining the light source distribution diagram of each subspace.
Wherein the light source distribution map comprises at least one light source.
The implementation and effect of this step can be understood with reference to step 108 in the embodiment shown in fig. 1, and will not be described in detail herein.
In this embodiment, in order to enable the preset network model to better perform light source layout on the plan, the preset network model is trained through a sample space with known light source distribution, so that a light source distribution diagram predicted based on the trained preset network model better accords with the actual situation or artificially laid light source distribution.
As shown in fig. 5, step 408 may include the following steps, based on the embodiment shown in fig. 4, described above:
Step 4081, preprocessing at least one sample space in the sample space set to obtain a sample image set and a supervision image set corresponding to the sample image set.
The sample image set comprises at least one sample image, the supervision image set comprises at least one supervision image, and the sample images are in one-to-one correspondence with the supervision images.
Optionally, the preprocessing of the sample space is similar to the structuring processing and the plane information extraction in the above embodiments, and the corresponding sample image and the supervisory image preset with the light source are obtained based on the sample space information corresponding to the sample space; a sample image and a supervisory image corresponding to each other can be obtained based on the same sample space data.
Step 4082, processing the sample image based on the preset network model to obtain a predicted light source distribution diagram.
And processing the sample image based on a preset network model to be trained, wherein the process of obtaining the predicted light source distribution diagram is the same as the application process.
Step 4083, training a preset network model based on the predicted illuminant distribution map and the network loss determined by the surveillance image.
In this embodiment, since the preset light source distribution map corresponds to the sample image one by one, the predicted light source distribution map corresponds to the monitor image one by one, the network loss can be determined based on the difference (e.g., pixel difference, etc.) between the predicted light source distribution map and the monitor image, and the preset network model can be trained by using the training method of the existing neural network based on the network loss, for example, the method of back gradient propagation, etc.; when the network loss reaches a preset condition, training of the preset neural network is completed, wherein the preset condition can include, but is not limited to: the network loss is smaller than a preset threshold value, the training times reach a preset number of times, and the like.
As shown in fig. 6, on the basis of the embodiment shown in fig. 5, step 4081 may include the following steps:
step 601, obtaining a sample space dataset based on sample space data for each sample space in the sample space set.
Wherein the sample space data includes space edge information in the sample space, at least one device information in the sample space, and at least one light source information in the sample space.
Step 602, performing imaging processing on each sample space data in the sample space data sets to obtain a sample image set and a supervision image set.
In this embodiment, the sample space data is subjected to the imaging processing based on the similar manner of the imaging processing in the above embodiment, which is different in that the sample space data further includes at least one light source, for example, a planar light source is abstract, a designer adds indoor surface light and generally irradiates from top to bottom, the light below the house top view is represented as a rectangle, so that the planar light source can be abstract as a 2D rectangle, the size of the rectangle represents the size of the planar light source, and the depth degree of the filling color of the rectangle represents the intensity of the light source; based on the information corresponding to the light source, the light source in the sample space is displayed in the supervision image during the imaging processing, the supervision image with the light source is used as supervision information, and a predicted light source distribution map predicted by the preset network model is made to approach to the supervision image, so that the light source distribution prediction capability of the preset network model is improved.
Optionally, based on the above embodiment, step 602 may further include:
carrying out structuring treatment on each sample space data in the sample space data set to obtain structured sample space data;
obtaining sample top view information corresponding to a sample space based on the structured sample space data;
extracting to obtain space edge information, at least one piece of equipment information and at least one piece of light source information based on sample top view information;
obtaining a sample image corresponding to the sample space based on the space edge information and at least one piece of equipment information;
and obtaining a supervision image corresponding to the sample space based on the space edge information, the at least one piece of equipment information and the at least one piece of light source information.
In this embodiment, before the data is structured, data cleaning may be further performed, and whether the structured sample space data obtained by the structured processing is standard or not is determined by an automation script (according to a preset rule), and the sample space data that does not conform to the preset rule is deleted from the sample space set, where the preset rule may include, but is not limited to: the data such as unreasonable house type (for example, the room has no door or window, etc.), unreasonable furniture position, unreasonable light source position, etc.
Optionally, on the basis of the above embodiment, step 4083 may include:
network loss is determined based on at least one predicted light source detection box included in the predicted light source profile and at least one supervising light source detection box included in the supervising image.
Wherein, the predictive light source detection frames are in one-to-one correspondence with the monitoring light source detection frames.
The preset network model is trained based on network loss.
In this embodiment, the structure of the preset network model may be obtained based on the structure of the existing neural network model, for example, a model is built by a variational self-encoder (VAE) model framework, the sample image generates hidden parameters through the encoding network, and then generates a predicted light source distribution map through the decoding network, which essentially is that the sample image can find the fitted characteristic parameters to generate the predicted light source distribution map through the preset network model; a network structure capable of realizing this function is applicable to the present embodiment; the network loss can be determined based on the difference value of the predicted light source distribution diagram and the regional pixel of the monitoring image, and the difference in a certain pixel range is allowed through certain parameter control, so that the fact that each pixel must be identical is not required, and the problem of model overfitting caused by too small set difference is avoided.
Optionally, in some optional embodiments, determining the network loss based on at least one predicted light source detection box included in the predicted light source profile and at least one supervising light source detection box included in the supervising image includes:
the first loss is determined based on an intersection ratio between each of the at least one predicted light source detection frames and the corresponding supervising light source detection frame.
Optionally, determining the difference in size and position between the light source detection frame and its corresponding supervisory control detection frame by predicting the ratio of intersection between the two, the smaller the ratio of intersection indicating the greater the difference between the two, the greater the corresponding first loss; a larger overlap ratio indicates a smaller difference between the two, and a corresponding smaller first loss.
In the color space dimension, a difference in brightness between each predicted light source detection box and the corresponding supervising light source detection box is determined, and a second loss is determined.
Optionally, each light source has light source Intensity information in addition to position and size information, optionally, when the light source Intensity is represented by different Brightness in the predicted light source distribution map, in order to determine the light source Intensity difference, the predicted light source distribution map and the supervisory image may be converted into a color space representing color Brightness for processing, for example, the preset light source distribution map is converted from RGB space to HSI color space, which is from human visual system, and describes color by Hue (Hue), saturation (Saturation or Chroma) and Brightness (Brightness or Brightness); the second penalty is determined by determining the luminance difference between each predicted light source detection box and the corresponding supervising light source detection box in the HSI color space, wherein the color space conversion may be achieved by any conversion technique in the prior art.
Based on the first loss and the second loss, a network loss is determined.
In this embodiment, the network loss may be determined based on a direct summation of the first loss and the second loss, or the network loss may be derived based on a weighted summation of the first loss and the second loss; the network loss obtained by combining the first loss and the second loss reflects the difference of the position, the size and the intensity information of the predicted light source and the light source in the monitoring image, and improves the prediction accuracy of the preset network model on the position, the size and the intensity of the light source.
Any of the light source layout methods provided by the embodiments of the present disclosure may be performed by any suitable device having data processing capabilities, including, but not limited to: terminal equipment, servers, etc. Alternatively, any of the light source arrangement methods provided by the embodiments of the present disclosure may be executed by a processor, such as the processor executing any of the light source arrangement methods mentioned by the embodiments of the present disclosure by invoking corresponding instructions stored in a memory. And will not be described in detail below.
Exemplary apparatus
Fig. 7 is a schematic diagram of an apparatus of a light source arrangement apparatus according to an exemplary embodiment of the present disclosure. As shown in fig. 7, the apparatus provided in this embodiment includes:
The data acquisition module 71 is configured to acquire spatial information data of the target space.
Wherein the target space comprises at least one subspace.
A sub-information module 72 for determining sub-information data for each sub-space based on the spatial information data.
A plan view module 73, configured to obtain a plan view of each subspace based on the sub-information data of each subspace.
The light source distribution module 74 is configured to process each plan view based on a preset network model, and determine a light source distribution map of each subspace.
Wherein the light source distribution map comprises at least one light source.
The light source layout device provided by the embodiment of the disclosure obtains spatial information data of a target space; wherein the target space comprises at least one subspace; determining sub-information data for each of the sub-spaces based on the spatial information data; obtaining a plan view of each subspace based on the subspace information data of each subspace; processing each plane map based on a preset network model respectively, and determining a light source distribution map of each subspace; wherein the light source distribution map comprises at least one light source; according to the embodiment, the plane diagram corresponding to the subspace is processed through the deep neural network, so that the light source distribution diagram with at least one light source arranged in the subspace can be directly obtained, automatic arrangement of the light source in the space is realized, the method is more suitable for industrial production, and the consumption of manpower and material resources caused by artificial arrangement of the light source in the prior art is overcome.
Optionally, the sub-information module 72 includes:
the structuring unit is used for carrying out structuring processing on the space information data to obtain structured space data;
and the data decomposition unit is used for carrying out decomposition processing on the structured space data and determining the sub-information data of each subspace.
Optionally, the plan view module 73 is specifically configured to obtain, for each subspace, plan view information corresponding to the subspace based on the structured sub-information data; extracting space edge information of the subspace and equipment information of at least one piece of equipment in the subspace based on the top view information; a plan view of the subspace is obtained based on the spatial edge information and the at least one device information.
In some optional embodiments, the apparatus provided in this embodiment further includes:
the model training module is used for training a preset network model based on the sample space set; wherein the set of sample spaces comprises at least one sample space.
Optionally, the model training module includes:
the preprocessing unit is used for preprocessing at least one sample space in the sample space set to obtain a sample image set and a supervision image set corresponding to the sample image set; the sample image set comprises at least one sample image, the supervision image set comprises at least one supervision image, and the sample images correspond to the supervision images one by one;
The light source prediction unit is used for processing the sample image based on a preset network model to obtain a predicted light source distribution diagram;
and the loss training unit is used for training a preset network model based on the predicted light source distribution diagram and the network loss determined by the supervision image.
Optionally, the preprocessing unit is specifically configured to obtain a sample space data set based on sample space data of each sample space in the sample space set; wherein the sample space data includes space edge information in the sample space, at least one device information in the sample space, and at least one light source information in the sample space; and carrying out imaging processing on each sample space data in the sample space data sets to obtain a sample image set and a supervision image set.
Optionally, the preprocessing unit is configured to perform a structuring process on each sample space data in the sample space data set to obtain structured sample space data when performing an imaging process on each sample space data in the sample space data set to obtain a sample image set and a supervision image set; obtaining sample top view information corresponding to a sample space based on the structured sample space data; extracting to obtain space edge information, at least one piece of equipment information and at least one piece of light source information based on sample top view information; obtaining a sample image corresponding to the sample space based on the space edge information and at least one piece of equipment information; and obtaining a supervision image corresponding to the sample space based on the space edge information, the at least one piece of equipment information and the at least one piece of light source information.
Optionally, the loss training unit is specifically configured to determine a network loss based on at least one predicted light source detection frame included in the predicted light source distribution map and at least one supervised light source detection frame included in the supervised image; wherein, the predicted light source detection frames correspond to the monitoring light source detection frames one by one; the preset network model is trained based on network loss.
Optionally, the loss training unit is configured to determine, when determining the network loss based on at least one predicted light source detection frame included in the predicted light source distribution map and at least one supervised light source detection frame included in the supervised image, a first loss based on an intersection ratio between each of the at least one predicted light source detection frames and the corresponding supervised light source detection frame; in the dimension of the color space, determining the brightness difference between each predicted light source detection frame and the corresponding monitoring light source detection frame, and determining a second loss; based on the first loss and the second loss, a network loss is determined.
Exemplary electronic device
Next, an electronic device according to an embodiment of the present disclosure is described with reference to fig. 8. The electronic device may be either or both of the first device and the second device, or a stand-alone device independent thereof, which may communicate with the first device and the second device to receive the acquired input signals therefrom.
Fig. 8 illustrates a block diagram of an electronic device according to an embodiment of the disclosure.
As shown in fig. 8, the electronic device 80 includes one or more processors 81 and memory 82.
Processor 81 may be a Central Processing Unit (CPU) or other form of processing unit having data processing and/or instruction execution capabilities and may control other components in electronic device 80 to perform desired functions.
The memory may store one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and/or nonvolatile memory. The volatile memory may include, for example, random Access Memory (RAM) and/or cache memory (cache), and the like. The non-volatile memory may include, for example, read Only Memory (ROM), hard disk, flash memory, and the like. One or more computer program products may be stored on the computer readable storage medium that can be run by a processor to implement the light source routing methods and/or other desired functions of the various embodiments of the present disclosure described above.
In one example, the electronic device 80 may further include: an input device 83 and an output device 84, which are interconnected by a bus system and/or other forms of connection mechanisms (not shown).
For example, when the electronic device is a first device or a second device, the input means 83 may be a microphone or a microphone array as described above for capturing an input signal of a sound source. When the electronic device is a stand-alone device, the input means 83 may be a communication network connector for receiving the acquired input signals from the first device and the second device.
In addition, the input device 83 may also include, for example, a keyboard, a mouse, and the like.
The output device 84 may output various information to the outside, including the determined distance information, direction information, and the like. The output means 84 may include, for example, a display, speakers, a printer, and a communication network and remote output devices connected thereto, etc.
Of course, only some of the components of the electronic device 80 relevant to the present disclosure are shown in fig. 8, with components such as buses, input/output interfaces, etc. omitted for simplicity. In addition, the electronic device 80 may include any other suitable components depending on the particular application.
Exemplary computer program product and computer readable storage Medium
In addition to the methods and apparatus described above, embodiments of the present disclosure may also be a computer program product comprising computer program instructions which, when executed by a processor, cause the processor to perform the steps in a light source layout method according to various embodiments of the present disclosure described in the above section of the present description.
The computer program product may write program code for performing the operations of embodiments of the present disclosure in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server.
Furthermore, embodiments of the present disclosure may also be a computer-readable storage medium, having stored thereon computer program instructions, which when executed by a processor, cause the processor to perform the steps in a light source arrangement method according to various embodiments of the present disclosure described in the above "exemplary method" section of the present disclosure.
The computer readable storage medium may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The basic principles of the present disclosure have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present disclosure are merely examples and not limiting, and these advantages, benefits, effects, etc. are not to be considered as necessarily possessed by the various embodiments of the present disclosure. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, since the disclosure is not necessarily limited to practice with the specific details described.
In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different manner from other embodiments, so that the same or similar parts between the embodiments are mutually referred to. For system embodiments, the description is relatively simple as it essentially corresponds to method embodiments, and reference should be made to the description of method embodiments for relevant points.
The block diagrams of the devices, apparatuses, devices, systems referred to in this disclosure are merely illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.
The methods and apparatus of the present disclosure may be implemented in a number of ways. For example, the methods and apparatus of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, firmware. The above-described sequence of steps for the method is for illustration only, and the steps of the method of the present disclosure are not limited to the sequence specifically described above unless specifically stated otherwise. Furthermore, in some embodiments, the present disclosure may also be implemented as programs recorded in a recording medium, the programs including machine-readable instructions for implementing the methods according to the present disclosure. Thus, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.
It is also noted that in the apparatus, devices and methods of the present disclosure, components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered equivalent to the present disclosure.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the disclosure. Thus, the present disclosure is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the disclosure to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.
Claims (10)
1. A light source arrangement method, comprising:
obtaining space information data of a target space; wherein the target space comprises at least one subspace;
determining sub-information data for each of the sub-spaces based on the spatial information data, the determining sub-information data for each of the sub-spaces based on the spatial information data comprising: carrying out structuring treatment on the space information data, and structuring the space information data into json data to obtain structured space data; decomposing the structured space data to determine sub-information data of each sub-space;
obtaining a plan view of each subspace based on the subspace information data of each subspace;
processing each plane map based on a preset network model respectively, and determining a light source distribution map of each subspace; wherein the light source distribution map comprises at least one light source.
2. The method of claim 1, wherein the obtaining a plan view of each subspace based on the subspace sub-information data comprises:
for each subspace, acquiring top view information corresponding to the subspace based on the structured subspace information data;
extracting space edge information of the subspace and equipment information of at least one piece of equipment in the subspace based on the top view information;
and obtaining a plan view of the subspace based on the space edge information and the at least one piece of equipment information.
3. The method according to any one of claims 1-2, further comprising, prior to processing each of the plan views separately based on a predetermined network model, determining a profile of the light source for each of the subspaces:
training the preset network model based on a sample space set; wherein the set of sample spaces comprises at least one sample space.
4. A method according to claim 3, wherein training the pre-set network model based on the set of sample spaces comprises:
preprocessing at least one sample space in the sample space set to obtain a sample image set and a supervision image set corresponding to the sample image set; the sample image set comprises at least one sample image, the supervision image set comprises at least one supervision image, and the sample images are in one-to-one correspondence with the supervision images;
Processing the sample image based on the preset network model to obtain a predicted light source distribution diagram;
and training the preset network model based on the predicted light source distribution diagram and the network loss determined by the supervision image.
5. The method of claim 4, wherein preprocessing at least one sample space in the set of sample spaces to obtain a set of sample images and a set of surveillance images corresponding to the set of sample images, comprises:
obtaining a sample space data set based on sample space data of each sample space in the sample space set; wherein the sample space data comprises space edge information in the sample space, at least one piece of equipment information in the sample space and at least one piece of light source information in the sample space;
and carrying out imaging processing on each sample space data in the sample space data set to obtain the sample image set and the supervision image set.
6. The method of claim 5, wherein said imaging each of said sample space data in said sample space data set to obtain said sample image set and said supervisory image set comprises:
Carrying out structuring treatment on each sample space data in the sample space data set to obtain structured sample space data;
obtaining sample top view information corresponding to the sample space based on the structured sample space data;
extracting the spatial edge information, the at least one piece of equipment information and the at least one piece of light source information based on the sample top view information;
obtaining a sample image corresponding to the sample space based on the space edge information and the at least one piece of equipment information;
and obtaining a supervision image corresponding to the sample space based on the space edge information, the at least one piece of equipment information and the at least one piece of light source information.
7. The method of claim 4, wherein the training the pre-set network model based on the predicted light source profile and the determined network loss of the surveillance image comprises:
determining the network loss based on at least one predicted light source detection box included in the predicted light source profile and at least one supervising light source detection box included in the supervising image; wherein the predicted light source detection frames correspond to the supervision light source detection frames one by one;
Training the preset network model based on the network loss.
8. The method of claim 7, wherein the determining the network loss based on at least one predicted light source detection box included in the predicted light source profile and at least one supervising light source detection box included in the supervising image comprises:
determining a first loss based on an intersection ratio between each of the at least one predicted light source detection frames and the corresponding supervising light source detection frame;
determining a brightness difference between each predicted light source detection frame and the corresponding supervision light source detection frame in a color space dimension, and determining a second loss;
the network loss is determined based on the first loss and the second loss.
9. An electronic device, comprising:
a memory for storing a computer program product;
a processor for executing the computer program product stored in the memory, and when executed, implementing the light source arrangement method of any of the preceding claims 1-8.
10. A computer readable storage medium having stored thereon computer program instructions, which when executed by a processor, implement the light source arrangement method of any of the preceding claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211123873.2A CN115438413B (en) | 2022-09-15 | 2022-09-15 | Light source arrangement method, electronic device, and computer-readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211123873.2A CN115438413B (en) | 2022-09-15 | 2022-09-15 | Light source arrangement method, electronic device, and computer-readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115438413A CN115438413A (en) | 2022-12-06 |
| CN115438413B true CN115438413B (en) | 2023-08-15 |
Family
ID=84248630
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211123873.2A Active CN115438413B (en) | 2022-09-15 | 2022-09-15 | Light source arrangement method, electronic device, and computer-readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115438413B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114329706A (en) * | 2021-12-23 | 2022-04-12 | 贝壳找房网(北京)信息技术有限公司 | Automatic wiring method and device |
| WO2022088901A1 (en) * | 2020-10-31 | 2022-05-05 | 华为技术有限公司 | Intelligent lamp switching method and system, and related device |
| CN114758068A (en) * | 2022-04-02 | 2022-07-15 | 北京地平线信息技术有限公司 | Training method and device of space geometric information estimation model |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR3037177B1 (en) * | 2015-06-08 | 2018-06-01 | Commissariat A L'energie Atomique Et Aux Energies Alternatives | IMAGE PROCESSING METHOD WITH SPECULARITIES |
| US11507713B2 (en) * | 2021-03-05 | 2022-11-22 | Flyreel, Inc. | Semi-supervised 3D indoor layout estimation from a single 360 degree panorama |
-
2022
- 2022-09-15 CN CN202211123873.2A patent/CN115438413B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022088901A1 (en) * | 2020-10-31 | 2022-05-05 | 华为技术有限公司 | Intelligent lamp switching method and system, and related device |
| CN114329706A (en) * | 2021-12-23 | 2022-04-12 | 贝壳找房网(北京)信息技术有限公司 | Automatic wiring method and device |
| CN114758068A (en) * | 2022-04-02 | 2022-07-15 | 北京地平线信息技术有限公司 | Training method and device of space geometric information estimation model |
Non-Patent Citations (1)
| Title |
|---|
| 分布式空间信息的对等协同计算机制研究;陈占龙;吴洁;谢忠;吴亮;;计算机应用研究(第07期);1-3节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115438413A (en) | 2022-12-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11520955B2 (en) | Structural design systems and methods using modeling and simulation for identifying equipment in floor plans | |
| US9129380B2 (en) | Background understanding in video data | |
| CN110717953B (en) | Coloring method and system for black-and-white pictures based on CNN-LSTM (computer-aided three-dimensional network-link) combination model | |
| CN110728333B (en) | Sunshine duration analysis method and device, electronic equipment and storage medium | |
| CN113673413A (en) | Method and device for examining architectural drawings, computer readable medium and electronic equipment | |
| CN111104879A (en) | Method and device for identifying house functions, readable storage medium and electronic equipment | |
| CN114972659A (en) | Method and system for converting two-dimensional drawing into three-dimensional model for drawing examination | |
| CN115438413B (en) | Light source arrangement method, electronic device, and computer-readable storage medium | |
| CN113762314B (en) | Firework detection method and device | |
| JPWO2020105146A1 (en) | Information processing equipment, control methods, and programs | |
| CN112991540B (en) | House model region segmentation method and device, storage medium and electronic equipment | |
| CN111986300B (en) | Rendering point determining method and device for house decoration, storage medium and electronic equipment | |
| Zhang et al. | Nonlinear prediction for Gaussian mixture image models | |
| US20230099463A1 (en) | Window, Door, and Opening Detection for 3D Floor Plans | |
| Voříšek et al. | Automated BIM Entity Reconstruction from Unstructured 3D Pointclouds | |
| CN116958269A (en) | External parameter calibration method of image pickup device, electronic device and storage medium | |
| Ipatov et al. | Segmentation Algorithm for the Evolutionary Biological Objects Images on a Complex Background | |
| CN115205157B (en) | Image processing method and system, electronic device and storage medium | |
| Qorbani et al. | A deep convolutional neural network based on U-Net to predict annual luminance maps | |
| Mukhopadhyay et al. | VR digital twin of office space with computer vision-based estimation of room occupancy and power consumption | |
| CN112989957A (en) | Safety monitoring method and system suitable for equipment cabinet | |
| Alawadhi et al. | Deep Learning Through Parametrically Generated Virtual Building Information Models for Real-World Object Recognition | |
| CN115620582A (en) | Industrial internet teaching training system and method and electronic equipment | |
| CN114241403A (en) | Method for detecting flexible object with indefinite direction, electronic device and storage medium | |
| Lehtimäki | GPU accelerated visual analytics tool for people flow tracking data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |