---
title: Dual-Domain Enhancement Module (DDE)
url: https://www.emergentmind.com/topics/dual-domain-enhancement-module-dde
type: topic
---

# Dual-Domain Enhancement Module (DDE)

Searching arXiv for recent papers using or closely related to “Dual-Domain Enhancement Module (DDE)” across imaging and signal-processing contexts.
Dual-Domain Enhancement Module (DDE) denotes a class of architectures that enhance a signal by jointly modeling two complementary representational domains rather than relying on a single feature space. In the cited literature, the paired domains are most often spatial and frequency/Fourier, but the same design principle also appears as color-gradient decomposition for low-light enhancement, spatial-spectral learning for underwater restoration, frequency-delay expansion for CSI prediction, and RGB–optical-flow differential fusion for video saliency. The term is used most explicitly in DEPF, where DDE combines Cross-Scale Wavelet Mamba (CSWM) and Fourier Details Recovery (FDR) to brighten low-light RGB remote-sensing images and recover texture details before multimodal fusion [2509.07327]. Closely related dual-domain modules appear in multimodal CT-PET synthesis [2606.13341], low-light enhancement [2309.08382; 2404.17400], underwater image enhancement [2504.19198; 2604.16266], infrared-visible fusion [2509.05000], UAV detection [2604.03176], and other signal-processing settings.

## 1. Conceptual scope and defining characteristics

The cited literature does not present DDE as a single canonical block. Instead, it presents a recurring architectural idea: decompose the input into two domains with complementary inductive biases, perform domain-specific enhancement, and then reconcile the enhanced representations for reconstruction, fusion, synthesis, or prediction. In DEPF, the two domains are low-frequency wavelet content and Fourier spectra; in DDE-GAN, they are image and frequency/measurement domains; in SS-UIE, they are spatial and spectral branches; and in DDNet, they are color and gradient domains rather than spatial and Fourier domains [2509.07327][2606.13341][2504.19198][2309.08382].

This literature suggests that DDE is better understood as a methodological family than as a fixed module specification. The family resemblance lies in coordinated enhancement across paired representations. One domain typically carries global illumination, structure, or low-frequency content, while the other carries edges, gradients, phase, high-frequency detail, or measurement-domain constraints. The dual-domain formulation is therefore used when single-domain processing is described as insufficient for preserving fine details, maintaining structural fidelity, handling degradation heterogeneity, or modeling long-range dependencies [2404.17400][2604.16266][2509.05000][2604.03176].

## 2. Architectural patterns

A common structural pattern is decomposition-enhancement-reconstruction. DEPF first applies N-level Haar 2D-DWT, enhances the low-frequency component through CSWM, reconstructs through inverse DWT, and then applies FDR in the Fourier domain [2509.07327]. DDNet divides enhancement into color enhancement and gradient enhancement through CEM and LoG-based GEM inside an encoder-decoder, followed by a final fusion decoder [2309.08382]. DFFN formalizes enhancement as two sequential phases—amplitude illumination and phase refinement—with information exchange through IFAM [2404.17400]. DDE-GAN uses paired generators in image and frequency domains and trains them through hierarchical dual-domain constraints [2606.13341].

| Work | Paired domains | Stated components |
|---|---|---|
| DEPF [2509.07327] | Wavelet low-frequency / Fourier spectrum | CSWM, FDR |
| DDNet [2309.08382] | Color / gradient | CEM, GEM |
| DFFN [2404.17400] | Amplitude / phase | DDAB, DDPB, IFAM |
| DDE-GAN [2606.13341] | Image / frequency or measurement | Dual generators, forward and inverse transforms |
| SS-UIE [2504.19198] | Spatial / spectral | MCSS, SWSA, SS-block |
| SFFNet [2604.03176] | Spatial edges / frequency edges | MDDC, DEIE |

Parallelism and hierarchy are both recurrent. SS-UIE splits features into two halves, sends one through SWSA and the other through MCSS, concatenates the outputs, and adds a residual connection [2504.19198]. Hero-Mamba processes the RGB image and FFT-derived spectral components in parallel and concatenates the outputs of Mamba-based SS2D blocks [2604.16266]. By contrast, DFFN is explicitly stage-wise: amplitude is learned first to restore brightness, and phase is learned second to refine details [2404.17400]. The choice between parallel and stage-wise organization reflects the specific role assigned to each domain: simultaneous complementarity in some settings, ordered correction in others.

## 3. Domain decompositions and operators

The operators used in DDE implementations vary substantially, but they all create a controlled separation of information types. In DEPF, wavelet decomposition is written as
$$
LL, [HL, LH, HH]_n, \ldots, [HL, LH, HH]_1 = DWT(I_v),
$$
after which CSWM enhances the low-frequency component via
$$
LL_e = LL \times \sum_i SSM(f_{seq}^i),
$$
and FDR derives amplitude and phase through FFT, enhances them with spectrum recovery networks, and reconstructs the output as
$$
I_{VE} = IFFT(AS_e, PS_e).
$$
This formulation makes the low-frequency branch responsible for global brightness and the Fourier branch responsible for texture-detail recovery [2509.07327].

DFFN uses Fourier amplitude and phase as explicit task variables rather than as auxiliary features. Its first-stage supervision target is
$$
\text{Target}_1 = \mathcal{F}^{-1}\left(\mathcal{A}(I_{gt}), \mathcal{P}(I_{low})\right),
$$
and the second stage receives
$$
I_{mix} = \mathcal{F}^{-1}\left(\mathcal{A}(O_A), \mathcal{P}(I_{low})\right).
$$
The resulting decomposition assigns brightness restoration to amplitude and detail refinement to phase [2404.17400].

Other DDE variants replace Fourier amplitude-phase analysis with alternative pairings. DDNet computes a LoG-based gradient map, concatenates it with the low-light image, and uses gradient-domain supervision for GEM alongside color-domain supervision for CEM [2309.08382]. SFFNet performs frequency-domain high-pass filtering, retains only frequencies above a threshold \(\alpha\), and modulates frequency magnitudes with a spatial edge-strength term \(S(i,j)\) before inverse transformation [2604.03176]. ChannelKAN generates a delay-domain CSI branch by applying IDFT to the frequency-domain CSI, then fuses the two streams downstream [2605.12553]. In each case, dual-domain enhancement depends not on a particular transform, but on an explicit partition between complementary signal attributes.

## 4. Optimization and supervision

DDE methods usually enforce consistency in both domains during training rather than relying solely on a final reconstruction loss. DDNet uses a joint loss composed of \(\mathcal{L}_{Lap}\), \(\mathcal{L}_{Coarse}\), and \(\mathcal{L}_{Final}\), with typical weights \(\omega_1 = 0.2\), \(\omega_2 = 0.2\), and \(\omega_3 = 0.6\). The first two terms are \(\ell_2\) losses for gradient and color supervision, and the final term is SSIM-based supervision for the fused output [2309.08382].

DDE-GAN organizes optimization into three stages. Stage 1 enforces intra-domain consistency in image and frequency domains. Stage 2 adds inter-domain consistency through forward and inverse projection operators. Stage 3 introduces rotational equivariance, expressed in the paper through constraints of the form
$$
\mathbf{T}_g\big( G^I_s(\mathbf{F}_s(\mathbf{x}_s)) \big) = G^I_s(\mathbf{F}_s(\mathbf{T}_g(\mathbf{x}_s))).
$$
This loss design makes the dual-domain formulation physically informed rather than merely feature-complementary [2606.13341].

Several works add domain-aware regularization terms that directly emphasize difficult frequencies or degraded subbands. SS-UIE introduces Frequency-Wise Loss (FWL) and combines it with pixel-domain supervision as
$$
\mathrm{L}_{total} = \mathrm{L}_1(\mathbf{I}_{gt}, \mathbf{I}_{pred}) + \lambda \cdot \mathrm{L}_{FWL}(\mathbf{F}_{gt}, \mathbf{F}_{pred}),
$$
where the frequency loss dynamically upweights harder frequency components [2504.19198]. GD\(^2\)Fusion uses intensity, texture, and color losses,
$$
L = \gamma L_{int} + \lambda L_{text} + \theta L_{color},
$$
with \(\gamma=5\), \(\lambda=5\), and \(\theta=6\), thereby coupling degradation suppression and fusion in an end-to-end objective rather than a decoupled pre-enhancement pipeline [2509.05000]. A plausible implication is that dual-domain enhancement is most effective when the training objective preserves the decomposition logic imposed by the architecture.

## 5. Applications and empirical record

DDE-style modules have been reported in enhancement, synthesis, fusion, detection, and prediction tasks.

| Work | Task | Reported result |
|---|---|---|
| DDNet [2309.08382] | UHD low-light enhancement | 4K \(3840 \times 2160\) at over 35 FPS; LOL PSNR 21.86, SSIM 0.832, NIQE 3.38 |
| DDE-GAN [2606.13341] | CT-PET synthesis | SSIM \(0.92 \pm 0.04\); PSNR \(28.12 \pm 1.92\) |
| Hero-Mamba [2604.16266] | Underwater enhancement on LSUI | PSNR 25.802; SSIM 0.913 |
| SFFNet-X [2604.03176] | UAV detection | 36.8 AP on VisDrone; 20.6 AP on UAVDT |
| ChannelKAN [2605.12553] | CSI prediction | NMSE 0.0265; SE 6.414; BER 0.00771 |

The empirical pattern is not restricted to image quality metrics. DDNet reports that object detection and scene segmentation improve on enhanced images in low-light ITS settings, and it is explicitly designed for real-time UHD surveillance [2309.08382]. DDE-GAN reports superior multimodal synthesis quality on the HECKTOR 2022 CT-PET dataset and attributes part of the gain to joint dual-domain learning with geometric equivariance [2606.13341]. In underwater enhancement, Hero-Mamba attributes its gains to parallel processing of spatial RGB and FFT components, while SS-UIE attributes its gains to adaptive spatial-spectral modeling with linear complexity [2604.16266][2504.19198]. In aerial detection, SFFNet reports that dual-domain edge enhancement improves AP over single-domain variants, while DEPF reports significant performance gains when CSWM and FDR are used together before multispectral fusion [2604.03176][2509.07327]. ChannelKAN’s ablation study further reports degradation from NMSE 0.0265 to 0.0318, from SE 6.414 to 6.385, and from BER 0.00771 to 0.00799 when dual-domain processing is removed [2605.12553].

## 6. Variants, misconceptions, and terminological boundaries

A common misconception is that DDE always denotes spatial-frequency image enhancement. The cited literature shows otherwise. DDNet defines the two domains as color and gradient [2309.08382]. The video salient object detection framework with Confidence-guided Adaptive Gate and Dual Differential Enhancement defines them as RGB and optical flow streams and enhances each stream by modeling feature differences,
$$
\mathcal{F}_e(X,Y) = \mathcal{F}_{conv}(Y-X)+X,
$$
before concatenation and fusion [2105.06714]. ChannelKAN uses frequency and delay domains rather than pixel and spectrum [2605.12553].

A second misconception is that DDE must be a standalone preprocessor. In practice, it may appear before a detector, inside a GAN, within a feature pyramid, or as part of a joint fusion pipeline. DEPF places DDE before RGB backbone extraction [2509.07327]. DDE-GAN embeds dual-domain learning inside both generator and discriminator losses [2606.13341]. GD\(^2\)Fusion distributes dual-domain processing across GFMSE and GSMAF branches and trains them jointly with the fusion network [2509.05000]. SFFNet integrates dual-domain edge enhancement into MDDC inside the detector backbone and neck [2604.03176].

The acronym itself is overloaded. In generative modeling, DDE can also mean Diffusion Domain Expansion, a method that coordinates multiple runs of a pre-trained diffusion model through a compact coordinator network; this usage is unrelated to dual-domain enhancement [2605.23275]. Comparable dual-domain logic also appears without the exact DDE label in MDPhD for time- and frequency-domain speech enhancement, D2Former for fully complex time-frequency speech enhancement, D\(^2\)Fusion for spatial-frequency deepfake detection, and DDSRNet for spatial-wavelet hyperspectral super-resolution [1812.08914][2302.11832][2503.17184][2512.09546]. This suggests that, in contemporary arXiv usage, “Dual-Domain Enhancement Module” functions less as a uniquely standardized module name than as a broadly reusable design principle for coupling complementary representational domains under task-specific supervision.

Source: https://www.emergentmind.com/topics/dual-domain-enhancement-module-dde