Weak supervision remote sensing image target detection method based on multi-modal pseudo-label guidance and adaptive fusion

Abstract

The invention relates to the technical field of image target detection, in particular to a weak supervision remote sensing image target detection method based on multi-mode pseudo-tag guidance and adaptive fusion. Performing dynamic fusion on the bimodal features through a cross-modal adaptive relation fusion module; and then a cross-modal pseudo-supervision label mapping module is utilized to generate and aggregate RGB and depth pseudo labels, and unified structure loss and depth dual loss are proposed to supervise and train a main branch and an auxiliary branch, so that the problems of insufficient utilization of multi-modal information, large pseudo label noise, insufficient fusion of global and local information and the like under weak supervision are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image target detection, and particularly relates to a weakly supervised remote sensing image target detection method based on multi-modal pseudo-label guidance and adaptive fusion. BACKGROUND

[0002] Weak Supervision-Camouflaged Object Detection (WS-COD) aims to reduce the dependence on a large amount of pixel-level labeled data and is suitable for detecting camouflaged objects in complex environments. Traditional camouflaged object detection methods require a large amount of high-quality labeling, which is labor-intensive and easily affected by subjective bias. Weak supervision methods use rough labeling (such as scribble-level, bounding box) to reduce labeling costs.

[0003] However, the high similarity of camouflaged objects and backgrounds in color, texture, and boundary brings great challenges to weak supervision. Camouflaged objects seamlessly blend in these aspects, making it difficult to capture subtle differences. Recent research has focused on improving weak supervision through adaptive mechanisms, multi-scale feature extraction, and boundary guidance. Other strategies include surpassing human vision, multi-modal fusion, and multi-task joint modeling. Despite the progress made, there are still challenges in effectively utilizing high-level visual features for weakly supervised sparse data classification.

[0004] In view of this, the present application provides a weakly supervised remote sensing image target detection method based on multi-modal pseudo-label guidance and adaptive fusion, aiming to realize accurate detection of remote sensing images under weak supervision. SUMMARY

[0005] The main purpose of the present application is to provide a weakly supervised remote sensing image target detection method based on multi-modal pseudo-label guidance and adaptive fusion, aiming to solve the problem of how to realize accurate detection of remote sensing images under weak supervision.

[0006] To achieve the above purpose, the present application provides a weakly supervised remote sensing image target detection method based on multi-modal pseudo-label guidance and adaptive fusion, characterized in that the method comprises the following steps:

[0007] S10, generating a depth map corresponding to the collected remote sensing image through a depth estimation model, and using a dual-stream encoder to extract RGB features and depth features in the depth map, respectively;

[0008] S20, fusing the RGB features and the depth features through a cross-modal adaptive relationship fusion module to obtain fused features;

[0009] S30, generating RGB pseudo-labels and depth feature pseudo-labels by using a cross-modal pseudo-supervised label mapping module, and aggregating to form pseudo RGB-D labels;

[0010] S40, based on the pseudo RGB-D labels, and using unified structure loss and depth double loss to supervise training of the main branch and the auxiliary branch;

[0011] S50, outputting the camouflage target detection result.

[0012] Optionally, the S20 comprises:

[0013] S21, performing convolution on the RGB feature and the depth feature to obtain a query vector Q, a key vector K and a value vector V;

[0014] S22, based on a cross-modal spatial view attention branch, calculating the attention score between each feature encoding block and performing position decoding to obtain a spatial view feature :

[0015]

[0016] wherein is a single head of the query vector, the key vector and the value vector respectively; is the number of heads, D represents the number and embedding dimension of pixel labels; e= is a scaling factor; represents a splicing operation of all heads, is an output projection matrix corresponding to the spatial view, used for fusing the heads;

[0017] S23, based on a cross-modal channel view attention branch, calculating the similarity between different feature channels to obtain a channel view feature :

[0018]

[0019] wherein is a scaling factor, is an output projection matrix corresponding to the channel view, used for fusing the heads;

[0020] S24, integrating the spatial view feature and the channel view feature by using learnable weights and to obtain a cross-modal attention depth feature output :

[0021]

[0022] wherein, and respectively represent the output features of the cross-modal spatial view attention branch and the cross-modal channel view attention branch, and are adjusted by learnable weights and respectively represent the output features of the cross-modal spatial view attention branch and the cross-modal channel view attention branch, and are adjusted by learnable weights and respectively represent the output features of the cross-modal spatial view attention branch and the cross-modal channel view attention branch, and are adjusted by learnable weights ;

[0023] S25, similarly, determines the cross-modal attention RGB feature output corresponding to the RGB feature :

[0024]

[0025] wherein represents the cross-modal attention output of the depth information propagating back to the RGB feature, and respectively represent the spatial attention output and the channel view attention output capturing the flow of depth information to the RGB feature, and two additional learnable weights and are introduced to adjust the contribution proportions of and ;

[0026] S26, each is enhanced with its corresponding cross-modal attention output , and then the enhanced feature maps are spliced to generate the final output fusion feature :

[0027]

[0028] In the formula, Cat(.,.) represents the splicing operation of the feature map, represents the original depth feature map extracted by the dual-stream encoder, represents the original RGB feature map extracted by the dual-stream encoder.

[0029] Optionally, the S30 comprises:

[0030] S31, the image is segmented into superpixel regions of a preset size; wherein if the number of foreground pixels in a superpixel in the superpixel region exceeds the number of background pixels and the number of foreground pixels is greater than 1, the superpixel is marked as foreground, otherwise, it is marked as background;

[0031] S32, by respectively diffusing the sparse scribbles Y into the superpixel regions generated from the RGB image and the superpixel regions generated from the depth map, if the foreground scribble pixels are more than the background, the region is marked as foreground, i.e. pseudo RGB label based on RGB texture boundary , otherwise marked as background i.e. pseudo depth label based on depth structure boundary ;

[0032] S33, using pixel adaptive refinement method to smooth and refine the pseudo RGB label and the pseudo depth label , and aggregate to form the pseudo RGB-D label :

[0033]

[0034] Optionally, the S40 comprises:

[0035] S41, using the unified structure loss to supervise the output of the main branch:

[0036]

[0037] wherein represents the binary cross-entropy loss, which is used to optimize the pixel-level accuracy, and focuses on the overall coincidence between the predicted region and the pseudo label as the real region; represents the prediction output of the i-th layer, represents the aggregated pseudo RGB-D label; represents the intersection over union loss, which is used to optimize the region overlap accuracy, and focuses on the overall coincidence between the predicted region and the pseudo label as the real region; is a structure similarity loss used to optimize the structure similarity of the image, which ensures that the predicted result is consistent with the pseudo label in structure. i and ∑ represent the multi-scale depth supervision index;

[0038] S42, using the depth double loss to supervise the auxiliary branch:

[0039]

[0040] wherein, is a binary cross-entropy loss, which is used to distinguish the pseudo object and its surrounding depth value by the pseudo depth label, is a depth smoothing loss, which is used to mask the background pixels by multiplying the pseudo RGB-D label with the depth prediction, to ensure the internal structure of the object to be smooth, and to reduce the high-frequency noise;

[0041] wherein:

[0042]

[0043]

[0044] wherein, represents the final prediction map generated by the deep branch, represents the pseudo-depth label represents the normal, p represents a pixel in the object region, is a neighboring pixel, cosine is the cosine similarity between two vectors, q represents a field pixel of pixel p, represents the gradient vector of the field of pixel p, represents the gradient information of pixel p on the prediction map ;

[0045] wherein:

[0046]

[0047]

[0048] wherein, and are Sobel operators, and represent the mask post-depth prediction values at the pixel points p and q;

[0049] S43, based on the unified structure loss and the depth double loss supervise the training of the main branch and the auxiliary branch.

[0050] The present application has at least the following beneficial effects:

[0051] Effectively solves the problems of insufficient utilization of multi-modal information, large pseudo-label noise, and insufficient fusion of global and local information under weak supervision, and realizes high-precision and high-efficiency saliency target detection on multiple data sets. BRIEF DESCRIPTION OF DRAWINGS

[0052] Figure 1 is a textual flowchart of a weakly supervised remote sensing image target detection method based on multi-modal pseudo-label guidance and adaptive fusion related to the embodiments of the present application;

[0053] Figure 2 is a graphical and textual flowchart of a weakly supervised remote sensing image target detection method based on multi-modal pseudo-label guidance and adaptive fusion related to the embodiments of the present application;

[0054] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0055] To better understand the above technical solutions, exemplary embodiments of this disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of this disclosure to those skilled in the art.

[0056] First Embodiment

[0057] Reference Figure 1 and Figure 2 The text and graphic flowcharts of a weakly supervised remote sensing image target detection method based on multimodal pseudo-label guidance and adaptive fusion are shown respectively. This embodiment provides a weakly supervised remote sensing image target detection method based on multimodal pseudo-label guidance and adaptive fusion, which includes the following steps:

[0058] S10, a depth map corresponding to the acquired remote sensing image is generated through a depth estimation model, and a dual-stream encoder is used to extract the RGB features and depth features from the depth map respectively;

[0059] In this embodiment, the collected remote sensing images are first input into the Depth Anything model in the field of depth estimation. By utilizing large-scale unlabeled data and the semantic prior knowledge of the pre-trained encoder, the model is forced to actively learn additional visual knowledge to obtain robust depth prediction capabilities and obtain the corresponding depth map. The RGB features and depth features in the depth map are then extracted respectively.

[0060] For example, we utilize the state-of-the-art technology in this field, DepthAnything V2, which provides us with an accurate depth map. Given an input RGB image of size [size missing], [details missing]. Here, H and W represent the spatial resolution of the image, respectively. Our goal is to leverage depth maps from multimodal data to facilitate the modeling of pseudo-depth labels, while combining spatial information to achieve more accurate segmentation of camouflaged objects. Initially, the input image I is fed into a monocular depth estimation network to generate a high-precision depth map D.

[0061] I and D are processed separately by a two-stream encoder to extract RGB and spatial features, respectively. Then, the RGB and spatial features are fused and decoded, serving as the main branch output Fout, designed to extract the boundary information of the camouflaged object. Simultaneously, the spatial features are decoded separately as the auxiliary branch output Dout to capture the position of the camouflaged object. Finally, pseudo-RGB labels and pseudo-depth labels are generated through graffiti annotation and the combined effect of I and D, converging into pseudo-RGB-D labels. These pseudo-RGB-D labels are used to supervise the entire network, while the pseudo-depth labels provide further auxiliary supervision to improve performance.

[0062] S20, the RGB features and the depth features are fused through the cross-modal adaptive relationship fusion module to obtain fused features;

[0063] In this embodiment, the input RGB feature map and depth feature map Convolution is performed to obtain the query vector Q, key vector K, and value vector V. Subsequently, positional encoding is applied to achieve block-level patch re-embedding; by recombining the query vector Q, key vector K, and value vector V, the attention mechanism operates at the block level, enabling each block to perform attention calculations independently. Finally, we utilize the multi-head self-attention mechanism in the transformer to calculate the attention score between each feature encoding block and perform positional decoding. It is evident that after interaction with different information sources, attention is more prominent near the edges of the camouflaged object. This can be expressed as:

[0064]

[0065] in and These are individual headers for the query, key, and value, respectively. It represents the number of heads, where D represents the number of pixel markers and the embedding dimension. e= It is the scaling factor. This represents the splicing operation of all heads. It is the output projection matrix of the corresponding spatial view, used to fuse these heads.

[0066] Cross-modal channel view attention is similar to cross-modal spatial view attention, the main difference being the omission of positional encoding and a shift in focus from spatial similarity to the similarity between independent feature channels. This method does not compare spatial locations within a feature sequence, but rather calculates the similarity between different feature channels, thereby capturing the complex relationships between channels in a multimodal feature map. The mathematical expression is:

[0067]

[0068] in It is the scaling factor. It is the output projection matrix corresponding to the channel view, used to fuse these headers.

[0069] Output features from cross-modal spatial view attention branch and cross-modal channel view attention branch and Through learnable weights and conduct Integration to generate cross-modal attention output This represents the propagation of RGB information to deeper features. We also aim to capture the propagation of deeper information to RGB features, denoted as... Therefore, we adjust the variables in the above formula accordingly: And introduce additional learnable weights. and To achieve this, we augment each original feature map with its corresponding cross-modal attention output, and then concatenate these augmented feature maps to generate the final output. Its expression is as follows:

[0070]

[0071] Summarized as follows:

[0072] S21, Convolve the RGB features and the depth features to obtain the query vector Q, the key vector K and the value vector V;

[0073] S22, based on the cross-modal spatial view attention branch, calculates the attention score between each feature coding block and performs position decoding to obtain the spatial view features. :

[0074]

[0075] in These are individual headers for the query vector, key vector, and value vector, respectively. It represents the number of heads, D represents the number of pixel markers and the embedding dimension; e= It is the scaling factor; This represents the splicing operation of all heads. It is the output projection matrix of the corresponding spatial view, used to fuse the various heads;

[0076] S23, Based on the cross-modal channel view attention branch, calculate the similarity between different feature channels to obtain channel view features. :

[0077]

[0078] in It is a scaling factor. It is the output projection matrix corresponding to the channel view, used for fusing the header;

[0079] S24, the spatial view features and the channel view features Through learnable weights and conduct Integration to generate cross-modal attention deep feature output :

[0080]

[0081] In the formula, and The output features of the cross-modal space view attention branch and the cross-modal channel view attention branch are represented by learnable weights. and Adjustment and The contribution ratio, generating cross-modal attention output. ;

[0082] S25, Similarly, determine the cross-modal attention RGB feature output corresponding to the RGB feature. :

[0083]

[0084] in The cross-modal attention output represents the propagation of depth information back to RGB features. and These represent the spatial attention output and channel view attention output, respectively, which capture depth information into the RGB feature flow. and Two additional learnable weights are used for adjustment and The proportion of contribution;

[0085] S26, each Its corresponding cross-modal attention output Enhancements are performed, and then these enhanced feature maps are concatenated to generate the final output fused feature. :

[0086]

[0087] In the formula, Cat(.,.) represents the feature map concatenation operation. This represents the original depth feature map extracted by the two-stream encoder. This represents the original RGB feature map extracted by the dual-stream encoder.

[0088] S30, uses the cross-modal pseudo-supervised label mapping module to generate RGB pseudo-labels and deep feature pseudo-labels, and aggregates them to form pseudo-RGB-D labels;

[0089] In this embodiment, a cross-modal pseudo-supervised label mapping module (CPLM) is used to generate and aggregate RGB and depth pseudo-labels to form pseudo-RGB-D labels. This step aims to create comprehensive pseudo-RGB-D labels by fusing pseudo-RGB and pseudo-depth labels to supervise the network's output. Furthermore, considering the unique challenges posed by camouflaged objects, pseudo-depth labels are introduced to refine the final prediction results of the depth branch, thereby achieving more accurate localization of camouflaged objects.

[0090] Specifically, and optionally, in this embodiment, we introduce a foreground-background selective expansion module to generate bimodal pseudo-labels. Specifically, the image is first segmented into superpixel regions of appropriate size. If the number of foreground pixels within a superpixel exceeds the number of background pixels, and the number of foreground pixels is greater than one, then the superpixel is marked as foreground. Otherwise, it is marked as background. Thus, by expanding the RGB and depth superpixel annotations to the Y region, corresponding pseudo-RGB labels are generated. and pseudo-depth tags To achieve a balance between efficiency and performance.

[0091] Furthermore, a pixel-adaptive thinning method, PAR, is used to smooth and thin the two pseudo-labels, ultimately generating an aggregated pseudo-RGB-D label. The pixel-adaptive thinning method defines a convolutional kernel that combines RGB and spatial information and is computed based on the neighboring locations of pixels. This kernel is applied to the average value through adaptive convolution iterations, effectively correcting for false labels.

[0092]

[0093] The decoded output from CARF is used as the main branch prediction, while the decoded output of the deep feature branch is used as the auxiliary prediction for supervision. A prediction head is applied and upsampled on the four decoded outputs of the main branch to achieve deep supervision and monitor the output of the initial layer in real time.

[0094] S40, based on the pseudo-RGB-D labels, and using a unified structure loss and depth dual loss to perform supervised training on the main branch and auxiliary branch;

[0095] In this embodiment, a unified structure loss is used to supervise the output of the main branch, and a deep dual loss is used to supervise the output of the main branch and auxiliary branches to complete the network supervised training.

[0096] Specifically, S41 adopts a unified structural loss. To monitor the output of the main branch:

[0097]

[0098] In the formula This represents the binary cross-entropy loss, used to optimize pixel-level accuracy, focusing on the overall overlap between the predicted region and the ground truth region (pseudo-label). This represents the predicted output of the i-th layer. This indicates the aggregation of pseudo-RGB-D tags. The intersection-union ratio loss is used to optimize the accuracy of region overlap, focusing on the overall degree of overlap between the predicted region and the real region (pseudo-label). This is a structural similarity loss used to optimize the structural similarity of images, ensuring that the predicted results are structurally consistent with the pseudo-labels. `i` and `∑` represent multi-scale deep supervision indices.

[0099] In this embodiment, the depth double loss includes two parts: (1) BCE loss, namely binary cross-entropy loss, which uses pseudo-depth labels to distinguish the depth values ​​of the disguised object from those around it; (2) depth smoothing loss, which uses pseudo-RGB-D labels to multiply with depth prediction to mask background pixels, ensure smooth internal structure of the object, and reduce high-frequency noise. The expression is:

[0100]

[0101] in:

[0102]

[0103]

[0104] In the formula, This represents the final prediction graph generated by the deep branch. Indicates pseudo-depth labels The normal vector is represented by p, and the pixel within the object region is represented by p. These are neighboring pixels, where cosine is the cosine similarity between two vectors, and q represents the neighborhood pixels of pixel p. This represents the gradient vector of the neighborhood of pixel p. Indicates that pixel p is in the prediction map Gradient information on;

[0105] in:

[0106]

[0107]

[0108] In the formula, and For the Sobel operator, and This represents the masked depth prediction values ​​at pixels p and q;

[0109] S50 outputs the detection results of camouflaged targets.

[0110] In the technical solution provided in this embodiment, the bimodal features are dynamically fused through the cross-modal adaptive relation fusion module (CARF); then, the RGB and depth pseudo-labels generated and aggregated by the cross-modal pseudo-supervised label mapping module (CPLM) are used to conduct supervised training on the main branch and auxiliary branch using a unified structural loss and depth dual loss, so as to solve the problems of insufficient utilization of multimodal information, large pseudo-label noise, and insufficient fusion of global and local information under weak supervision.

[0111] Verification Implementation Examples

[0112] Based on the first embodiment, this embodiment verifies the detection performance of the weakly supervised remote sensing image target detection method based on multimodal pseudo-label guidance and adaptive fusion proposed in this application:

[0113] First, in the data preparation phase, we used the S-COD training set, introduced by He et al., which marked the first dataset specifically tailored for COD tasks under a weakly supervised framework. For the test set, we used four widely accepted datasets: CHAMELEON, CAMO, COD10K, and NC4K.

[0114] The initialization of model parameters specifically includes the parameters of the dual-stream encoder, CARF module, and CPLM module.

[0115] Then, PVT-v2-B2 was used as the backbone network, initialized with pre-trained weights, while RGB images and depth maps used separate backbone networks. Under this setup, the overall computational complexity was 182.13 GFLOPs, and the total number of parameters was 53.33M. For comprehensive comparison, we also provided an SPMCNet variant using a ResNet50 backbone. The input image size was resized to 352×352 pixels, and the number of superpixels was set to 100. To reduce overfitting, random data augmentation was employed, including scaling, flipping, rotation, and color enhancement. The model was trained for 50 epochs using the Nadam optimizer with a batch size of 8.

[0116] In this embodiment, we use four commonly used indicators for evaluation, including structural indicators. Average E-measurement Adaptive F-measurement and mean absolute error

[0117] The following is a quantitative comparison of the model across four benchmarks. "F" represents a fully supervised method, "U" represents an unsupervised method, "S" represents a scribble-supervised method, and "P" represents a point-supervised method. The best results for both scribble-supervised and point-supervised methods are highlighted in bold.

[0118] Table 1. Quantitative comparison of the model across four benchmark tests

[0119]

[0120] Table 1 (continued)

[0121]

[0122] Experimental results show that the method proposed in this invention outperforms multiple methods on multiple datasets, verifying its effectiveness and superiority.

[0123] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0124] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A weakly supervised remote sensing image target detection method based on multimodal pseudo-label guidance and adaptive fusion, characterized in that, The method includes the following steps: S10, a depth map corresponding to the acquired remote sensing image is generated through a depth estimation model, and a dual-stream encoder is used to extract the RGB features and depth features from the depth map respectively; S20, the RGB features and the depth features are fused through the cross-modal adaptive relationship fusion module to obtain fused features; S30, uses the cross-modal pseudo-supervised label mapping module to generate RGB pseudo-labels and deep feature pseudo-labels, and aggregates them to form pseudo-RGB-D labels; S40, based on the pseudo-RGB-D labels, and using a unified structure loss and depth dual loss to perform supervised training on the main branch and auxiliary branch; S50 outputs the detection results of camouflaged targets.

2. The method as described in claim 1, characterized in that, S20 includes: S21, Convolve the RGB features and the depth features to obtain the query vector Q, the key vector K and the value vector V; S22, based on the cross-modal spatial view attention branch, calculates the attention score between each feature coding block and performs position decoding to obtain the spatial view features. : ； in These are the individual headers of the query vector, key vector, and value vector, respectively; e= It is a scaling factor. It represents the number of heads, and D represents the number of pixel markers and the embedding dimension; This represents the splicing operation of all heads. It is the output projection matrix of the corresponding spatial view, used to fuse the various heads; S23, Based on the cross-modal channel view attention branch, calculate the similarity between different feature channels to obtain channel view features. : ； in It is a scaling factor. It is the output projection matrix corresponding to the channel view, used for fusing the header; S24, the spatial view features and the channel view features Through learnable weights and conduct Integration to generate cross-modal attention deep feature output : ； In the formula, and The output features of the cross-modal space view attention branch and the cross-modal channel view attention branch are represented by learnable weights. and Adjustment and The contribution ratio, generating cross-modal attention output. ; S25, Similarly, determine the cross-modal attention RGB feature output corresponding to the RGB feature. : ； in The cross-modal attention output represents the propagation of depth information back to RGB features. and These represent the spatial attention output and channel view attention output, respectively, which capture depth information into the RGB feature flow. and Two additional learnable weights are used for adjustment and The proportion of contribution; S26, each Its corresponding cross-modal attention output Enhancements are performed, and then these enhanced feature maps are concatenated to generate the final output fused feature. : ； In the formula, Cat(.,.) represents the feature map concatenation operation. This represents the original depth feature map extracted by the two-stream encoder. This represents the original RGB feature map extracted by the dual-stream encoder.

3. The method as described in claim 1, characterized in that, S30 includes: S31, the image is divided into superpixel regions of a preset size; wherein, if the number of foreground pixels in a superpixel region exceeds the number of background pixels, and the number of foreground pixels is greater than 1, then the superpixel is marked as foreground; otherwise, it is marked as background. S32, by spreading the sparse graffiti annotation Y to the superpixel regions generated from the RGB image and the superpixel regions generated from the depth map respectively, the region with more foreground graffiti pixels than background is labeled as the foreground, i.e., pseudo-RGB labeling based on RGB texture boundaries ( Otherwise, it is marked as background, i.e., pseudo-depth label based on depth structure boundary. ); S33, employing a pixel-adaptive thinning method on the pseudo-RGB label. and the pseudo-depth label Smooth and refine the data, then aggregate it to form pseudo-RGB-D labels. : 。 4. The method as described in claim 1, characterized in that, S40 includes: S41, employing a unified structural loss To monitor the output of the main branch: ； In the formula This represents the binary cross-entropy loss, used to optimize pixel-level accuracy, focusing on the overall overlap between the predicted region and the pseudo-labels that serve as the real region. This represents the predicted output of the i-th layer. This indicates the aggregation of pseudo-RGB-D tags; The intersection-union ratio loss is used to optimize the accuracy of region overlap, focusing on the overall degree of overlap between the predicted region and the pseudo-labels that are the real regions. This is a structural similarity loss used to optimize the structural similarity of images, ensuring that the predicted results are structurally consistent with the pseudo-labels. `i` and `∑` represent multi-scale deep supervision indexes; S42, using deep double loss To supervise the auxiliary branch: ； In the formula, The binary cross-entropy loss is used to distinguish camouflaged objects from their surrounding depth values ​​in pseudo-depth labels. The depth smoothing loss is used to mask background pixels by multiplying pseudo-RGB-D labels with depth prediction, ensuring smooth internal structures of objects and reducing high-frequency noise. in: ； ； In the formula, This represents the final prediction graph generated by the deep branch. Indicates pseudo-depth labels; The normal vector is represented by p, and the pixel within the object region is represented by p. These are neighboring pixels, where cosine is the cosine similarity between two vectors, and q represents the neighborhood pixels of pixel p. This represents the gradient vector of the neighborhood of pixel p. Indicates that pixel p is in the prediction map Gradient information on; in: ； ； In the formula, and For the Sobel operator, and This represents the masked depth prediction values ​​at pixels p and q; S43, Based on Unified Structural Loss Both depth loss Supervised training is performed on the main branch and auxiliary branches.

Images