Variable-granularity self-adaptive coding method for smoke and dust disaster environment

Abstract

The invention relates to the technical field of image coding, in particular to a variable-granularity self-adaptive coding method for a smoke and dust disaster environment, which comprises the following steps: performing dust infrared transmissivity dynamic compensation based on the obtained current dust concentration and visibility of a smoke and dust disaster environment scene, and calculating a transmissivity compensation coefficient; synchronously acquiring a visible light image and an SWIR image of the smoke and dust disaster environment scene, and performing gain compensation on the SWIR image based on the transmissivity compensation coefficient to obtain a compensated SWIR image; performing feature extraction and fusion on the visible light image and the compensated SWIR image to obtain fusion features, and restoring the fusion features into a fusion image; selecting the size of a DCT block in combination with the local texture complexity, and performing DCT on the fused image to obtain a group of DCT coefficients; and applying quantization parameter threshold constraint to the abnormal high-frequency DCT coefficient, and carrying out variable granularity DCT quantization on the fused image to obtain entropy coding.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of image coding technology, and in particular to a variable granularity adaptive coding method for smoke and dust disaster environments. Background Art

[0002] In industrial and mining enterprises like coal mines and chemical plants, sudden fires and explosions are often accompanied by large amounts of smoke and dust. For example, if a conveyor belt catches fire in a mine, the smoke produced contains CO2, carbon particles, and aerosols. If the fire spreads to adjacent mining areas, it can cause ventilation systems to reverse course. The CO concentration will diffuse throughout the mine within two hours, triggering Mie scattering. The resulting smoke blurs visible light images, rendering monitoring systems ineffective. Solid particles such as coal dust and rock dust in mining disasters have an infrared absorption rate of up to 60%, attenuating human body temperature signals by over 50%. Infrared thermal imaging equipment cannot identify the heat source of trapped individuals, hindering accurate search and rescue efforts. If a fire causes an explosion, the combined contamination of smoke and dust can interfere with multimodal data sources such as visible light, infrared, and gas sensors. High-temperature smoke plumes can cause infrared sensors to misidentify the heat source, while dust coverage prevents visible light cameras from capturing information about the collapse of critical structures, complicating rescue efforts.

[0003] At the same time, after a mining explosion, visibility in the mine and factory areas plummets, dropping from 10 meters to 0.5 meters within one second. PM2.5 concentrations instantly exceed 1000 μg / m³. The CO concentration in a fire rises from 0.1% to 2% in just 30 seconds, and the oxygen concentration plummets from 21% to below 12%. This results in a sudden change in the combustion state and large fluctuations in gas concentrations. Furthermore, the temperature in the fire source area can reach over 800°C, but the temperature gradient in the smoke diffusion path exhibits a nonlinear distribution, with a temperature difference of 200°C per meter. This sudden change in parameters and nonlinear correlations further complicate the changes in dynamic environmental parameters.

[0004] Smoke and dust during disasters can pose the following challenges to a company's safe production.

[0005] First, it renders enterprise safety monitoring and early warning systems ineffective. In smoky and dusty environments, the false alarm rate of traditional cameras increases by 40%, with dust clouds mistakenly identified as flames, resulting in fire warning delays exceeding five minutes. Dust adhesion can reduce the sensitivity of gas sensors by 70%, leading to methane detection errors of up to ±15%.

[0006] Second, it increases safety risks for equipment and personnel. Dust accumulation can cause equipment to short-circuit and explode, increasing the failure rate of electrical equipment in coal mines by three times.

[0007] Third, production interruption and economic losses. A single dust pollution incident can cause a production line to shut down for 8 to 24 hours, resulting in direct losses exceeding one million yuan.

[0008] Fourth, disaster relief efforts are more complicated. When smoke reduces visibility to less than one meter, rescue robots can experience navigation errors exceeding three meters. In dusty environments, the effective detection range of infrared life detectors is reduced to 30% of its original range, hindering rescue route planning. High-temperature smoke plumes exceeding 300°C can shorten the effectiveness of protective clothing to just 10 minutes, causing rescue workers to become unconscious after just 15 minutes of exposure and dramatically increasing the risk of CO poisoning. When dust concentrations exceed 300g / m³, static sparks can trigger explosions, and ventilation system disruptions can increase the spread of fire by 50%.

[0009] In smoke and dust disaster environments, quickly and accurately identifying key disaster features is crucial. However, traditional image coding methods lack adaptability to complex environments. In low-visibility environments like smoke and dust, they cannot effectively identify key disaster features (such as infrared heat sources and human outlines), resulting in data transmission distortion. Traditional image coding methods use fixed encoding modes and fail to dynamically optimize based on real-time meteorological parameters (visibility, particulate matter concentration). This leads to insufficient penetration, resulting in wasted bandwidth and loss of critical information. Summary of the Invention

[0010] In order to solve the above technical problems, the embodiments of the present application propose a variable granularity adaptive coding method for smoke and dust disaster environments, aiming to utilize dust infrared transmittance dynamic compensation technology and multi-spectral fusion coding technology to improve the heat source recognition rate, significantly reduce the bandwidth required for the encoding process, improve the encoding speed, and be able to combine real-time meteorological parameters to achieve dynamic optimization and enhancement, which well meets the high real-time and high-precision emergency response requirements in smoke and dust disaster environment scenarios.

[0011] In order to achieve the above-mentioned purpose, an embodiment of the present application proposes a variable granularity adaptive encoding method for smoke and dust disaster environments, and the method includes the following steps: obtaining the current dust concentration and visibility of the smoke and dust disaster environment scene, and when the dust concentration is greater than a preset first dust concentration threshold or the visibility is less than a preset first visibility threshold, dynamically compensating the dust infrared transmittance based on the dust concentration and visibility, and calculating the transmittance compensation coefficient; synchronously starting a visible light camera and a SWIR camera to shoot the smoke and dust disaster environment scene, obtaining a visible light image and a SWIR image of the smoke and dust disaster environment scene, and performing gain compensation on the SWIR image based on the transmittance compensation coefficient to obtain a compensated SWIR image; respectively Feature extraction is performed on the visible light image and the compensated SWIR image to obtain visible light features and SWIR features. The visible light features and SWIR features are then fused based on the attention mechanism and the band weight matrix to obtain fused features, which are then restored to a fused image. The DCT block size is selected based on the local texture complexity based on the Zernike moment, and the fused image is subjected to DCT transformation based on the selected DCT block size to obtain a set of DCT coefficients. Abnormal high-frequency coefficients are detected based on the mapping relationship between dust particle size and DCT coefficients, and quantization parameter threshold constraints are imposed on the abnormal high-frequency coefficients. The fused image is subjected to variable-granularity DCT quantization based on the quantization parameter and the quantization parameter threshold constraint to obtain entropy coding of the smoke and dust disaster environment scene.

[0012] To achieve the above objectives, an embodiment of the present application further proposes a variable granularity adaptive coding system for smoke and dust disaster environments, the method comprising: a dust concentration sensor, a visibility observer, a transmittance compensation coefficient calculation module, an image acquisition and compensation module, a visible light camera, a SWIR camera, a feature extraction and feature fusion module, a DCT transformation module, a dust suppression module for variable granularity coding, and a DCT quantization module; the dust concentration sensor is used to obtain the current dust concentration and the current dust particle size of the smoke and dust disaster environment scene; the visibility observer is used to obtain the current visibility of the smoke and dust disaster environment scene; the transmittance compensation coefficient calculation module is used to detect whether the dust concentration is greater than a preset first dust concentration threshold and detect whether the visibility is less than a preset first visibility threshold. When the dust concentration is greater than the preset first dust concentration threshold or the visibility is less than the preset first visibility threshold, dynamic compensation for the infrared transmittance of dust is performed based on the dust concentration and visibility to calculate the transmittance compensation coefficient; the image acquisition and compensation module is used to synchronously start the visible light camera and the SWIR camera to detect the smoke and dust. The smoke and dust disaster environment scene is photographed to obtain visible light images and SWIR images of the smoke and dust disaster environment scene, and the SWIR image is gain compensated based on the transmittance compensation coefficient to obtain the compensated SWIR image; the feature extraction and feature fusion module is used to extract features from the visible light image and the compensated SWIR image respectively to obtain visible light features and SWIR features, and the visible light features and SWIR features are fused based on the attention mechanism and the band weight matrix to obtain fused features, and then restored to a fused image; the DCT transformation module is used to select the DCT block size in combination with the local texture complexity based on the Zernike moment, and perform DCT transformation on the fused image based on the selected DCT block size to obtain a set of DCT coefficients; the variable granularity coded dust suppression module is used to detect abnormal high-frequency coefficients based on the mapping relationship between dust particle size and DCT coefficients, and impose quantization parameter threshold constraints on the abnormal high-frequency coefficients; the DCT quantization module is used to perform variable granularity DCT quantization on the fused image based on the quantization parameter and the quantization parameter threshold constraint to obtain entropy coding of the smoke and dust disaster environment scene.

[0013] In order to achieve the above-mentioned purpose, an embodiment of the present application also provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute a variable granularity adaptive coding method for smoke and dust disaster environments as described above.

[0014] In order to achieve the above-mentioned purpose, an embodiment of the present application further provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it can implement a variable granularity adaptive coding method for smoke and dust disaster environments as described above.

[0015] This application proposes a variable granularity adaptive coding method for smoke and dust disaster environments. It makes targeted improvements to traditional image coding methods for smoke and dust disaster environments from three aspects: dynamic compensation of dust infrared transmittance, multi-spectral fusion coding, and dust suppression of variable granularity coding. The traditional technology for compensating for dust concentration can only adjust the global brightness, while this application dynamically compensates for dust infrared transmittance based on dust concentration and visibility, so that the heat source characteristics remain recognizable in the smoke and dust. The compensated heat source signal-to-noise ratio is greatly improved, which improves the detection efficiency of human targets. In terms of multi-spectral utilization, this application performs gain compensation on SWIR images, performs feature extraction, fusion and restoration based on visible light images and compensated SWIR images, and obtains a fused image that retains the highly penetrating band. As the basic data for subsequent encoding, it greatly enhances the adaptability to smoke and dust disaster environments. In terms of suppression mechanism, traditional image coding methods usually adopt the most basic fixed low-pass filtering technology, which lacks flexibility. This application detects abnormal high-frequency coefficients based on the mapping relationship between dust particle size and DCT coefficients, imposes quantization parameter threshold constraints on abnormal high-frequency coefficients, and realizes adaptive filtering, which greatly enhances the penetration ability of smoke and dust. In summary, this application effectively improves the heat source recognition rate, significantly reduces the bandwidth required for the image coding process, improves the coding speed, and can achieve dynamic optimization and enhancement in combination with real-time meteorological parameters, which well meets the high real-time and high-precision emergency response requirements in smoke and dust disaster environment scenarios.

[0016] Optionally, the current dust concentration and visibility of the smoke and dust disaster environment scene are obtained. When the dust concentration is greater than a preset first dust concentration threshold or the visibility is less than a preset first visibility threshold, dynamic compensation of dust infrared transmittance is performed based on the dust concentration and visibility, including: obtaining the current dust concentration and visibility of the smoke and dust disaster environment scene, detecting whether the dust concentration is greater than the preset first dust concentration threshold, and detecting whether the visibility is less than the preset first visibility threshold; if the dust concentration is detected to be greater than the preset first dust concentration threshold, or the visibility is detected to be less than the preset first visibility threshold, dynamic compensation of dust infrared transmittance is performed based on the dust concentration and visibility; if the dust concentration is detected to be not greater than the preset first dust concentration threshold, and the visibility is detected to be not less than the preset first visibility threshold, conventional encoding is performed.

[0017] Optionally, dynamic compensation of dust infrared transmittance is performed based on dust concentration and visibility, and a transmittance compensation coefficient is calculated, including: The rate adjustment factor is determined based on dust concentration and visibility using the following formula: ; ; ; ; in, and Respectively represent the current dust concentration and visibility, 、 and represent the preset first dust concentration threshold, second dust concentration threshold and third dust concentration threshold respectively, 、 and represent the preset first visibility threshold, second visibility threshold and third visibility threshold respectively, Indicates the determined rate adjustment coefficient, 、 and Respectively represent the preset first rate adjustment coefficient, second rate adjustment coefficient and third rate adjustment coefficient; The transmittance compensation coefficient is calculated using the following formula based on the bitrate adjustment coefficient, the preset number of channels, and the depth of field information of the smoke and dust disaster environment scene: ; in, Indicates the preset number of channels, Represents the depth of field information of smoke and dust disaster environment scene, Represents the calculated transmittance compensation coefficient.

[0018] Optionally, gain compensation is performed on the SWIR image based on the transmittance compensation coefficient using the following formula to obtain a compensated SWIR image: ; in, represents a SWIR image, represents the compensated SWIR image.

[0019] Optionally, feature extraction is performed on the visible light image and the compensated SWIR image respectively to obtain visible light features and SWIR features, and feature fusion is performed on the visible light features and SWIR features based on the attention mechanism and the band weight matrix to obtain fused features, including: The visible light image and compensated SWIR image Input to the pre-trained dual-branch feature extraction model, the visible light branch of the dual-branch feature extraction model is used to extract the Perform feature extraction to obtain visible light features , the thermal infrared branch of the dual-branch feature extraction model Perform feature extraction to obtain thermal infrared features Among them, the network structure of the visible light branch is ResNet-18, and the network structure of the thermal infrared branch is the improved U-Net; Based on the band weight matrix, the following formula is used right Perform weighted reorganization to obtain the thermal infrared characteristics after band weighting : ; ; ; in, express point The corresponding band, express point The corresponding band weight; Through the following formula, based on the attention mechanism, As the query vector, As key vector and value vector, and Perform feature fusion to obtain fusion features: ; ; ; in, 、 、 Represent query vector, key vector, and value vector respectively, Indicates dimension, upper right corner mark Indicates the transpose operation. represents the Softmax function, Indicates fusion features.

[0020] Optionally, the DCT block size is selected in conjunction with local texture complexity based on Zernike moments, including: The Zernike moment is calculated based on the visible light image using the following formula: ; in, Indicates size The visible light image of express A point in The pixel value of Indicates a point The corresponding repetition rate is of order Zernike polynomials, The calculated repetition rate is of order Zernike moments; Based on the following formula, , calculate texture complexity: ; in, represents the total order, Indicates the calculated texture complexity; Based on the following formula, , choose the DCT block size: ; ; in, and are respectively the preset first texture complexity threshold and the second texture complexity threshold, Indicates the selected DCT block size.

[0021] Optionally, abnormal high-frequency coefficients are detected based on the mapping relationship between dust particle size and DCT coefficients, and quantization parameter threshold constraints are imposed on the abnormal high-frequency coefficients, including: determining the abnormal high-frequency threshold based on the mapping relationship between dust particle size and DCT coefficients; comparing each DCT coefficient with the abnormal high-frequency threshold, determining DCT coefficients less than the abnormal high-frequency threshold as normal coefficients, and determining DCT coefficients greater than or equal to the abnormal high-frequency threshold as abnormal high-frequency coefficients; for normal coefficients, assigning preset conventional quantization parameters; for abnormal high-frequency coefficients, applying quantization parameter threshold constraints on the basis of conventional quantization parameters, obtaining the constrained quantization parameters and assigning them to the abnormal high-frequency coefficients. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the related technologies, the following is a brief introduction to the drawings required for use in the embodiments of the present application or the description of the related technologies. Obviously, the following drawings are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. The drawings described here are only used to explain the present application and are not used to limit the present application.

[0023] Figure 1 This is a flow chart of a variable granularity adaptive encoding method for smoke and dust disaster environments provided in one embodiment of the present application; Figure 2 This is a flowchart of selecting conventional coding or infrared feature enhanced coding based on dust concentration and visibility provided in one embodiment of the present application; Figure 3 is a schematic diagram of the feature extraction, fusion, and restoration process provided in one embodiment of the present application; Figure 4 is an entropy coding heat map provided in one embodiment of the present application; Figure 5 This is a schematic structural diagram of a variable granularity adaptive coding system for smoke and dust disaster environments provided in another embodiment of the present application; Figure 6 It is a structural diagram of an electronic device provided in another embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the embodiments of the present application will be described in detail below with reference to the accompanying drawings. Those skilled in the art will appreciate that in the embodiments of the present application, many technical details are proposed to enable the reader to better understand. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed in the present application can be implemented. The division of the following embodiments is for convenience of description and should not constitute any limitation on the specific implementation of the present application. The following embodiments can be combined with each other and referenced to each other under the premise of no contradiction.

[0025] One embodiment of the present application proposes a variable-granularity adaptive encoding method for smoke and dust disaster environments, which is applied to an electronic device, where the electronic device can be a terminal or a server. This embodiment and the following embodiments are described using a server as an example. The following describes the implementation details of the variable-granularity adaptive encoding method for smoke and dust disaster environments proposed in this embodiment. The following content is only provided for ease of understanding and is not required for the implementation of this solution.

[0026] The specific process of the variable granularity adaptive coding method for smoke and dust disaster environments proposed in this embodiment can be as follows: Figure 1 Shown, including: Step 11: Obtain the current dust concentration and visibility of the smoke and dust disaster environment scene. When the dust concentration is greater than the preset first dust concentration threshold or the visibility is less than the preset first visibility threshold, perform dynamic compensation for the dust infrared transmittance based on the dust concentration and visibility, and calculate the transmittance compensation coefficient.

[0027] In the specific implementation, facing the smoke and dust disaster environment scenario, the server first needs to obtain the current dust concentration through the dust concentration sensor, and obtain the current visibility through the visibility observer, and then perform threshold comparison on the current dust concentration and the current visibility respectively. When the current dust concentration is greater than the preset first dust concentration threshold, or the current visibility is less than the preset first visibility threshold, the dust infrared transmittance is dynamically compensated based on the dust concentration and visibility, and the transmittance compensation coefficient is calculated.

[0028] In one example, after obtaining the current dust concentration and current visibility of the smoke and dust disaster environment scene, the server detects whether the current dust concentration is greater than the preset first dust concentration threshold, and detects whether the current visibility is less than the preset first visibility threshold. If it is detected that the current dust concentration is greater than the preset first dust concentration threshold, or if it is detected that the current visibility is less than the preset first visibility threshold, then the infrared transmittance of the dust is dynamically compensated based on the current dust concentration and the current visibility, and the transmittance compensation coefficient is calculated, that is, infrared feature enhancement coding is selected. If it is detected that the current dust concentration is not greater than the preset first dust concentration threshold, and if it is detected that the current visibility is not less than the preset first visibility threshold, then there is no need to perform dynamic compensation for the infrared transmittance of the dust, and conventional coding can be performed directly. This design effectively saves computing resources and can improve the efficiency of coding to a certain extent.

[0029] In one example, the process of selecting conventional coding or infrared signature enhanced coding based on dust concentration and visibility is as follows: Figure 2 As shown, the first dust concentration threshold is 500μg / m³, and the first visibility threshold is 5m. If the current dust concentration is detected to be greater than 500μg / m³, or the current visibility is detected to be less than 5m, infrared feature enhancement coding is selected; if the current dust concentration is detected to be no greater than 500μg / m³, and the current visibility is detected to be no less than 5m, conventional coding is selected.

[0030] In one example, when the server dynamically compensates for dust infrared transmittance based on dust concentration and visibility and calculates the transmittance compensation coefficient, it first determines the bitrate adjustment coefficient based on the dust concentration and visibility using the following formula: ; ; ; ; in, and Respectively represent the current dust concentration and visibility, 、 and represent the preset first dust concentration threshold, second dust concentration threshold and third dust concentration threshold respectively, 、 and represent the preset first visibility threshold, second visibility threshold and third visibility threshold respectively, Indicates the determined rate adjustment coefficient, 、 and They represent the preset first rate adjustment coefficient, second rate adjustment coefficient and third rate adjustment coefficient respectively.

[0031] In one example, 500μg / m³, 750 μg / m³, It is 1000μg / m³.

[0032] In one example, 5m, 3m, is 1m.

[0033] In one example, is 1.2, is 1.6, is 2.0.

[0034] In one example, after calculating the transmittance compensation coefficient, the server can calculate the transmittance compensation coefficient based on the calculated bit rate adjustment coefficient, the preset number of channels, and the depth of field information of the smoke and dust disaster environment scene: ; in, Indicates the preset number of channels, Represents the depth of field information of smoke and dust disaster environment scene, Represents the calculated transmittance compensation coefficient.

[0035] Step 12: Synchronously start the visible light camera and the SWIR camera to shoot the smoke and dust disaster environment scene, obtain a visible light image and a SWIR image of the smoke and dust disaster environment scene, and perform gain compensation on the SWIR image based on the transmittance compensation coefficient to obtain a compensated SWIR image.

[0036] In practice, after determining the need for infrared feature enhancement encoding, the server simultaneously activates the visible light camera and SWIR camera to capture the smoke and dust disaster environment scene, obtaining visible light and SWIR images of the smoke and dust disaster environment scene. The SWIR image is then gain-compensated based on the transmittance compensation coefficient to produce a compensated SWIR image. Heat sources in the compensated SWIR image are identifiable, and the heat source signal-to-noise ratio can be improved by 22dB.

[0037] In one example, a visible light camera can use the SONY IMX585, and a SWIR camera can use an InGaAs sensor. The two cameras are synchronized by hardware, and the time deviation is less than 1ms.

[0038] In an example, the server may perform gain compensation on the SWIR image based on the transmittance compensation coefficient using the following formula to obtain a compensated SWIR image: ; in, represents a SWIR image, represents the compensated SWIR image.

[0039] In step 13, feature extraction is performed on the visible light image and the compensated SWIR image respectively to obtain visible light features and SWIR features, and feature fusion of the visible light features and SWIR features is performed based on the attention mechanism and the band weight matrix to obtain fused features, and the fused images are restored.

[0040] In the specific implementation, after obtaining the compensated SWIR image, the server can extract features from the visible light image and the compensated SWIR image respectively to obtain visible light features and SWIR features, and then perform feature fusion of the visible light features and SWIR features based on the attention mechanism and the band weight matrix to obtain fused features, and finally restore the fused features to a fused image.

[0041] In one example, the feature extraction, fusion, and restoration process can be as follows Figure 3 As shown. Feature extraction is implemented by a dual-branch feature extraction model, which consists of a visible light branch and a thermal infrared branch. The server converts the visible light image and compensated SWIR image Input into the pre-trained dual-branch feature extraction model, and use the visible light branch of the dual-branch feature extraction model to extract the Perform feature extraction to obtain visible light features , using the thermal infrared branch of the dual-branch feature extraction model Perform feature extraction to obtain thermal infrared features Among them, the network structure of the visible light branch is ResNet-18, and the network structure of the thermal infrared branch is the improved U-Net.

[0042] In one example, the feature fusion process is implemented by a fusion model, which consists of a weighted recombination unit and an attention unit. The weighted recombination unit is based on the band weight matrix Thermal infrared characteristics Perform band weighted recombination to obtain the thermal infrared characteristics after band weighting The attention unit is responsible for the attention mechanism. As the query vector, As key vector and value vector, and Perform feature fusion to obtain fusion features .

[0043] In one example, the weighted recombination unit is based on the band weight matrix by the following formula right Perform weighted reorganization to obtain the thermal infrared characteristics after band weighting : ; ; ; in, express point The corresponding band, express point The corresponding band weight.

[0044] In one example, the attention unit uses the following formula to convert As the query vector, As key vector and value vector, and Perform feature fusion to obtain fusion features : ; ; ; in, 、 、 Represent query vector, key vector, and value vector respectively, Indicates dimension, upper right corner mark Indicates the transpose operation. Represents the Softmax function.

[0045] In one example, the feature restoration process is implemented based on the restoration model, which is based on the fusion feature Perform feature restoration to restore the fused image .

[0046] Step 14: Select a DCT block size based on the local texture complexity based on the Zernike moment, and perform DCT transformation on the fused image based on the selected DCT block size to obtain a set of DCT coefficients.

[0047] In the specific implementation, the server obtains the fused image After that, the DCT block size can be selected in combination with the local texture complexity of the Zernike moment, and then the fused image can be fused based on the selected DCT block size. Perform DCT transformation to obtain a set of DCT coefficients.

[0048] In one example, the server calculates the Zernike moment based on the visible light image using the following formula: ; in, Indicates size The visible light image of express A point in The pixel value of Indicates a point The corresponding repetition rate is of order Zernike polynomials, The calculated repetition rate is of order Zernike moments.

[0049] In the calculation Then, based on the following formula , calculate texture complexity: ; in, represents the total order, Indicates the calculated texture complexity.

[0050] Finally, based on the following formula , choose the DCT block size: ; ; in, and are respectively the preset first texture complexity threshold and the second texture complexity threshold, Indicates the selected DCT block size.

[0051] Step 15: Detect abnormal high-frequency coefficients based on the mapping relationship between dust particle size and DCT coefficients, impose quantization parameter threshold constraints on the abnormal high-frequency coefficients, perform variable-granularity DCT quantization on the fused image based on the quantization parameter and the quantization parameter threshold constraints, and obtain entropy coding of the smoke and dust disaster environment scene.

[0052] In practice, after completing the DCT transform, DCT quantization is ready. The server first detects abnormal high-frequency coefficients based on the mapping relationship between dust particle size and DCT coefficients, applies a quantization parameter threshold constraint to these coefficients, and then performs variable-granularity DCT quantization on the fused image based on the quantization parameter and the quantization parameter threshold constraint, thereby obtaining the entropy encoding of the smoke and dust disaster environment scene.

[0053] In one example, the entropy coding heat map of a smoke and dust disaster environment scene is as follows: Figure 4 shown.

[0054] In one example, for the DCT coefficients obtained by DCT transformation, the server first determines the abnormal high-frequency threshold based on the preset mapping relationship between the dust particle size and the DCT coefficient. Subsequently, the server compares each DCT coefficient with the abnormal high-frequency threshold in turn, determines the DCT coefficients less than the abnormal high-frequency threshold as normal coefficients, and determines the DCT coefficients greater than or equal to the abnormal high-frequency threshold as abnormal high-frequency coefficients. For normal coefficients, the server assigns a preset conventional quantization parameter. For abnormal high-frequency coefficients, the server needs to impose a quantization parameter threshold constraint on the basis of the conventional quantization parameter, obtain the constrained quantization parameter, and assign it to the abnormal high-frequency coefficient.

[0055] In one example, the dust particle size is detected by a dust concentration sensor, and the minimum and maximum values ​​of the DCT coefficients corresponding to the dust particle size are recorded in the preset mapping relationship between the dust particle size and the DCT coefficient. The server uses the maximum value of the DCT coefficient corresponding to the dust particle size as the abnormal high-frequency threshold.

[0056] This embodiment proposes a variable granularity adaptive coding method for smoke and dust disaster environments. This method improves traditional image coding methods specifically for smoke and dust disaster environments from three aspects: dynamic compensation for dust infrared transmittance, multispectral fusion coding, and dust suppression through variable granularity coding. Traditional techniques for compensating for dust concentration only adjust global brightness. This embodiment dynamically compensates for dust infrared transmittance based on dust concentration and visibility, allowing heat source features to remain recognizable in smoke and dust. The compensated heat source signal-to-noise ratio is significantly improved, enhancing the detection efficiency of human targets. Regarding multispectral utilization, this embodiment performs gain compensation on SWIR images and performs feature extraction, fusion, and restoration based on visible light images and compensated SWIR images. This produces a fused image that retains highly penetrating bands. This fused image serves as the basis for subsequent coding, significantly enhancing adaptability to smoke and dust disaster environments. In terms of suppression mechanism, traditional image coding methods usually adopt the most basic fixed low-pass filtering technology, which lacks flexibility. This embodiment detects abnormal high-frequency coefficients based on the mapping relationship between dust particle size and DCT coefficients, and imposes quantization parameter threshold constraints on abnormal high-frequency coefficients to achieve adaptive filtering, which greatly enhances the penetration ability of smoke and dust. In summary, the variable granularity adaptive coding method for smoke and dust disaster environments proposed in this embodiment effectively improves the heat source recognition rate, significantly reduces the bandwidth required for the image coding process, improves the coding speed, and can be combined with real-time meteorological parameters to achieve dynamic optimization and enhancement, which well meets the high real-time and high-precision emergency response requirements in smoke and dust disaster environment scenarios.

[0057] The steps of the various methods above are divided for clarity of description only. During implementation, they can be combined into a single step, or some steps can be broken down into multiple steps. As long as they contain the same logical relationships, they are all within the scope of protection of this application. Adding minor modifications or introducing minor design changes to the algorithm or process, but not changing the core design of the algorithm or process, are also within the scope of protection of this application.

[0058] Another embodiment of the present application proposes a variable granularity adaptive coding system for smoke and dust disaster environments. The details of the variable granularity adaptive coding system for smoke and dust disaster environments proposed in this embodiment are described in detail below. The following content is only the implementation details provided for easy understanding and is not necessary for the implementation of this example. Figure 51 is a schematic diagram of the structure of a variable granularity adaptive coding system for smoke and dust disaster environments proposed in this embodiment, including: a dust concentration sensor 21, a visibility observer 22, a transmittance compensation coefficient calculation module 31, an image acquisition and compensation module 32, a visible light camera 23, a SWIR camera 24, a feature extraction and feature fusion module 33, a DCT transformation module 34, a variable granularity coding dust suppression module 35 and a DCT quantization module 36.

[0059] The dust concentration sensor 21 is used to obtain the current dust concentration and the current dust particle size in the smoke and dust disaster environment scene.

[0060] The visibility observer 22 is used to obtain the current visibility of the smoke and dust disaster environment scene.

[0061] The transmittance compensation coefficient calculation module 31 is used to detect whether the dust concentration is greater than a preset first dust concentration threshold and whether the visibility is less than a preset first visibility threshold. When the dust concentration is greater than the preset first dust concentration threshold or the visibility is less than the preset first visibility threshold, dynamic compensation for the dust infrared transmittance is performed based on the dust concentration and visibility to calculate the transmittance compensation coefficient.

[0062] The image acquisition and compensation module 32 is used to synchronously start the visible light camera 23 and the SWIR camera 24 to shoot the smoke and dust disaster environment scene, obtain the visible light image and SWIR image of the smoke and dust disaster environment scene, and perform gain compensation on the SWIR image based on the transmittance compensation coefficient generated by the transmittance compensation coefficient calculation module 31 to obtain the compensated SWIR image.

[0063] The feature extraction and feature fusion module 33 is used to extract features from the visible light image and the compensated SWIR image respectively to obtain visible light features and SWIR features, perform feature fusion on the visible light features and SWIR features based on the attention mechanism and the band weight matrix to obtain fused features, and restore them to a fused image.

[0064] The DCT transformation module 34 is configured to select a DCT block size based on the local texture complexity based on the Zernike moment, and perform DCT transformation on the fused image based on the selected DCT block size to obtain a set of DCT coefficients.

[0065] The variable granularity coding dust suppression module 35 is used to detect abnormal high-frequency coefficients based on the mapping relationship between dust particle size and DCT coefficients, and impose quantization parameter threshold constraints on the abnormal high-frequency coefficients.

[0066] The DCT quantization module 36 is used to perform DCT quantization of a variable granularity on the fused image based on a quantization parameter and a quantization parameter threshold constraint to obtain entropy coding of the smoke and dust disaster environment scene.

[0067] It is not difficult to find that this embodiment is a system embodiment corresponding to the above-mentioned method embodiment, and this embodiment can be implemented in conjunction with the above-mentioned method embodiment. The relevant technical details and technical effects mentioned in the above-mentioned method embodiment are still valid in this embodiment, and to reduce repetition, they are not repeated here. Accordingly, the relevant technical details mentioned in this embodiment can also be applied to the above-mentioned method embodiment.

[0068] It is worth mentioning that all modules and modules involved in this embodiment are logical modules. In actual applications, a logical unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. In addition, in order to highlight the innovation of this application, this embodiment does not introduce units that are not closely related to solving the technical problems proposed by this application, but this does not mean that other units do not exist in this embodiment.

[0069] Another embodiment of the present application provides an electronic device, such as Figure 6 As shown, it includes: at least one processor 41; and a memory 42 communicatively connected to the at least one processor 41; wherein the memory 42 stores instructions that can be executed by the at least one processor 41, and the instructions are executed by the at least one processor 41 to enable the at least one processor 41 to execute a variable granularity adaptive coding method for smoke and dust disaster environments as described in the above method embodiment.

[0070] The memory and processor are connected using a bus, which includes any number of interconnected buses and bridges. The bus connects various circuits of one or more processors and memories. The bus can also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits. These are all well known in the art and therefore will not be described further in this article. The bus interface provides an interface between the bus and the transceiver. The transceiver can be a single component or multiple components, such as multiple receivers and transmitters, providing a unit for communicating with various other devices on a transmission medium. Data processed by the processor is transmitted on a wireless medium via an antenna. Furthermore, the antenna also receives data and transmits it to the processor.

[0071] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory can be used to store data used by the processor when performing operations.

[0072] Another embodiment of the present application proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it can implement a variable granularity adaptive coding method for smoke and dust disaster environments as described in the above method embodiment.

[0073] That is, those skilled in the art will understand that all or part of the steps in the above-described method embodiments can be implemented by instructing the relevant hardware through a program, which is stored in a storage medium and includes a number of instructions for causing a device (such as a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps in the method embodiments of the present application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a mobile hard drive, a read-only memory, a random access memory, a magnetic disk, or an optical disk.

[0074] Those skilled in the art will appreciate that the above embodiments are specific embodiments for implementing the present application. In actual applications, various modifications may be made to the embodiments in form and detail without departing from the spirit and scope of the present application. Those skilled in the art will appreciate that improvements and modifications may be made without departing from the principles of the present application, and such improvements and modifications are also considered to be within the scope of protection of the present application.

Claims

1. A variable granularity adaptive coding method for smoke and dust disaster environments, characterized by: The method comprises: Obtain the current dust concentration and visibility of the smoke and dust disaster environment scene. When the dust concentration is greater than a preset first dust concentration threshold or the visibility is less than a preset first visibility threshold, perform dynamic compensation for the dust infrared transmittance based on the dust concentration and visibility, and calculate a transmittance compensation coefficient. Synchronously start the visible light camera and the SWIR camera to shoot the smoke and dust disaster environment scene, obtain visible light images and SWIR images of the smoke and dust disaster environment scene, and perform gain compensation on the SWIR image based on the transmittance compensation coefficient to obtain a compensated SWIR image; Feature extraction is performed on the visible light image and the compensated SWIR image respectively to obtain visible light features and SWIR features. Feature fusion of the visible light features and SWIR features is performed based on the attention mechanism and the band weight matrix to obtain fused features, and then the fused image is restored. The DCT block size is selected based on the local texture complexity of the Zernike moment, and the fused image is subjected to DCT transformation based on the selected DCT block size to obtain a set of DCT coefficients; Based on the mapping relationship between dust particle size and DCT coefficient, abnormal high-frequency coefficients are detected, and quantization parameter threshold constraints are imposed on the abnormal high-frequency coefficients. Based on the quantization parameter and quantization parameter threshold constraints, variable-granularity DCT quantization is performed on the fused image to obtain the entropy coding of the smoke and dust disaster environment scene.

2. The variable granularity adaptive coding method for smoke and dust disaster environments according to claim 1 is characterized in that: Obtain the current dust concentration and visibility of the smoke and dust disaster environment scene. When the dust concentration is greater than a preset first dust concentration threshold or the visibility is less than a preset first visibility threshold, perform dynamic compensation for dust infrared transmittance based on the dust concentration and visibility, including: Obtain the current dust concentration and visibility of the smoke and dust disaster environment scene, detect whether the dust concentration is greater than a preset first dust concentration threshold, and detect whether the visibility is less than a preset first visibility threshold; If the dust concentration is detected to be greater than a preset first dust concentration threshold, or if the visibility is detected to be less than a preset first visibility threshold, dynamic compensation of the dust infrared transmittance is performed based on the dust concentration and visibility; If the dust concentration is detected to be no greater than a preset first dust concentration threshold, and the visibility is detected to be no less than a preset first visibility threshold, conventional encoding is performed.

3. The variable granularity adaptive coding method for smoke and dust disaster environments according to claim 2 is characterized in that: Dynamic compensation of dust infrared transmittance is performed based on dust concentration and visibility, and the transmittance compensation coefficient is calculated, including: The rate adjustment factor is determined based on dust concentration and visibility using the following formula: ； ； ； ； in, and Respectively represent the current dust concentration and visibility, 、 and represent the preset first dust concentration threshold, second dust concentration threshold and third dust concentration threshold respectively, 、 and represent the preset first visibility threshold, second visibility threshold and third visibility threshold respectively, Indicates the determined rate adjustment coefficient, 、 and Respectively represent the preset first rate adjustment coefficient, second rate adjustment coefficient and third rate adjustment coefficient; The transmittance compensation coefficient is calculated using the following formula based on the bitrate adjustment coefficient, the preset number of channels, and the depth of field information of the smoke and dust disaster environment scene: ； in, Indicates the preset number of channels, Represents the depth of field information of smoke and dust disaster environment scene, Represents the calculated transmittance compensation coefficient.

4. The variable granularity adaptive coding method for smoke and dust disaster environments according to claim 3 is characterized in that: The SWIR image is gain compensated based on the transmittance compensation coefficient using the following formula to obtain the compensated SWIR image: ； in, represents a SWIR image, represents the compensated SWIR image.

5. The variable granularity adaptive coding method for smoke and dust disaster environments according to claim 4 is characterized in that: Feature extraction is performed on the visible light image and the compensated SWIR image respectively to obtain visible light features and SWIR features. The visible light features and SWIR features are fused based on the attention mechanism and the band weight matrix to obtain fused features, including: The visible light image and compensated SWIR image Input to the pre-trained dual-branch feature extraction model, the visible light branch of the dual-branch feature extraction model is used to extract the Perform feature extraction to obtain visible light features , the thermal infrared branch of the dual-branch feature extraction model Perform feature extraction to obtain thermal infrared features Among them, the network structure of the visible light branch is ResNet-18, and the network structure of the thermal infrared branch is the improved U-Net; Based on the band weight matrix, the following formula is used right Perform weighted reorganization to obtain the thermal infrared characteristics after band weighting : ； ； ； in, express point The corresponding band, express point The corresponding band weight; Through the following formula, based on the attention mechanism, As the query vector, As key vector and value vector, and Perform feature fusion to obtain fusion features: ； ； ； in, 、 、 Represent query vector, key vector, and value vector respectively, Indicates dimension, upper right corner mark Indicates the transpose operation. represents the Softmax function, Indicates fusion features.

6. A variable granularity adaptive coding method for smoke and dust disaster environments according to any one of claims 1 to 5, characterized in that: The DCT block size is selected in combination with the local texture complexity based on Zernike moments, including: The Zernike moment is calculated based on the visible light image using the following formula: ； in, Indicates size The visible light image of express A point in The pixel value of Indicates a point The corresponding repetition rate is of order Zernike polynomials, The calculated repetition rate is of order Zernike moments; Based on the following formula, , calculate texture complexity: ； in, represents the total order, Indicates the calculated texture complexity; Based on the following formula, , choose the DCT block size: ； ； in, and are respectively the preset first texture complexity threshold and the second texture complexity threshold, Indicates the selected DCT block size.

7. The variable granularity adaptive coding method for smoke and dust disaster environments according to claim 6 is characterized in that: Abnormal high-frequency coefficients are detected based on the mapping relationship between dust particle size and DCT coefficients, and quantization parameter threshold constraints are imposed on abnormal high-frequency coefficients, including: Based on the mapping relationship between dust particle size and DCT coefficient, the abnormal high-frequency threshold is determined; Comparing each DCT coefficient with an abnormal high frequency threshold, determining a DCT coefficient less than the abnormal high frequency threshold as a normal coefficient, and determining a DCT coefficient greater than or equal to the abnormal high frequency threshold as an abnormal high frequency coefficient; For normal coefficients, a preset conventional quantization parameter is assigned; For abnormal high-frequency coefficients, a quantization parameter threshold constraint is imposed on the basis of a conventional quantization parameter to obtain a constrained quantization parameter and allocate it to the abnormal high-frequency coefficients.

8. A variable granularity adaptive coding system for smoke and dust disaster environments, characterized by: The system includes: a dust concentration sensor, a visibility observer, a transmittance compensation coefficient calculation module, an image acquisition and compensation module, a visible light camera, a SWIR camera, a feature extraction and feature fusion module, a DCT transformation module, a variable granularity coding dust suppression module and a DCT quantization module; Dust concentration sensor, used to obtain the current dust concentration and current dust particle size in smoke and dust disaster environment scenes; Visibility observation instrument, used to obtain the current visibility of smoke and dust disaster environment scenes; a transmittance compensation coefficient calculation module, configured to detect whether the dust concentration is greater than a preset first dust concentration threshold and whether the visibility is less than a preset first visibility threshold; and, if the dust concentration is greater than the preset first dust concentration threshold or the visibility is less than the preset first visibility threshold, dynamically compensate for the dust infrared transmittance based on the dust concentration and visibility to calculate a transmittance compensation coefficient; The image acquisition and compensation module is used to synchronously start the visible light camera and the SWIR camera to shoot the smoke and dust disaster environment scene, obtain the visible light image and the SWIR image of the smoke and dust disaster environment scene, and perform gain compensation on the SWIR image based on the transmittance compensation coefficient to obtain the compensated SWIR image; The feature extraction and feature fusion module is used to extract features from the visible light image and the compensated SWIR image respectively to obtain visible light features and SWIR features, perform feature fusion on the visible light features and SWIR features based on the attention mechanism and the band weight matrix to obtain fused features, and restore them to a fused image; A DCT transformation module is used to select a DCT block size based on the local texture complexity based on the Zernike moment, and perform DCT transformation on the fused image based on the selected DCT block size to obtain a set of DCT coefficients; The dust suppression module with variable granularity coding is used to detect abnormal high-frequency coefficients based on the mapping relationship between dust particle size and DCT coefficients, and impose quantization parameter threshold constraints on abnormal high-frequency coefficients; The DCT quantization module is used to perform DCT quantization of the fused image with variable granularity based on the quantization parameter and the quantization parameter threshold constraint to obtain the entropy coding of the smoke and dust disaster environment scene.

9. An electronic device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the variable granularity adaptive coding method for smoke and dust disaster environments as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it can implement a variable granularity adaptive encoding method for smoke and dust disaster environments according to any one of claims 1 to 7.

Images