Low-light image enhancement （LLIE） is a key challenge in computational photography and computer vision，with critical applications ranging from autonomous navigation and night surveillance to computational imaging. Although existing methods in sRGB，HSV，and YCbCr color spaces have shown effectiveness，they face an irreconcilable trade-off between preserving color fidelity and suppressing luminance artifacts，often introducing chromatic shifts or halo distortions during the enhancement process. Recent studies focus on color space optimization. YCbCr-based methods perform shallow luminance processing （e.g.，linear filtering），causing an imbalance between chrominance denoising and detail preservation. Accordingly，the performance of these methods is capped. HSV space strategies decouple the hue and saturation components but cause color discontinuities due to polar coordinate quantization. This study proposes a novel lightweight LLIE network to address these limitations. This network fully leverages the channel decoupling properties of the HVI color space，minimizing the mutual interference between luminance information and color structural information during enhancement.

We propose GCDINet，an end-to-end framework leveraging HVI color space’s physical decoupling. The architecture comprises： 1） a dual-branch structure that processes value-intensity and hue components independently via the GCFE module，which utilizes dual GELU activations and element-wise multiplication to generate spatial attention masks； 2） a dual-path adaptive feature calibration （DAFC） structure that nonlinearly modulates illumination and chrominance features before reconstruction； 3） a composite loss combining HVI-space and sRGB-space constraints. The novel HVI color space decouples images into luminance and color structure maps at the original resolution，allowing the dual-branch network to process luminance and color structure information independently. This mechanism avoids interference from processing in the same feature stream and provides independent branch information for the GCFE module. A DAFC structure nonlinearly modulates the illumination features from the luminance branch and the chrominance vectors from the color branch before feature reconstruction. This approach guides information reconstruction during decoding，suppressing color fidelity loss and luminance artifact amplification caused by information loss. Although the dual-branch structure and gated convolutions decouple luminance and color structure information for a refined feature enhancement，uneven channel weight distribution during feature fusion can result in the loss of key luminance and chrominance information. During training，we optimize using a composite loss function that combines L1 intensity loss，structural similarity index measure （SSIM） loss，and edge-aware regularization. The model is implemented in PyTorch with the Adam optimizer（

Comprehensive evaluations on the LOLv1，LOLv2-Real，and LOLv2-Synthetic datasets demonstrate the state-of-the-art performance of GCDINet. On the LOLv1 dataset，our method achieves the best results in terms of SSIM and learned perceptual image patch similarity （LPIPS） and is only 0.76 dB short of the best method in peak signal-to-noise ratio （PSNR）. On the LOLv1-Real dataset，our method attains optimal results in PSNR and SSIM and is only 0.006 short of the best method in LPIPS. Experimental results on the LOLv2-Synthetic dataset demonstrate that the proposed method achieves a PSNR of 25.698 dB，an SSIM of 0.935，and an LPIPS score of 0.047. These metrics show improvements of 0.028 dB，5.3%，and 20.3% compared with the state-of-the-art LLIE method Retinexformer. Furthermore，the computational complexity is 7.94 GFLOPs，which is 50% lower than that of Retinexformer，achieving an effective balance between speed and performance compared to the existing methods. Ablation studies show that the PSNR，SSIM，and LPIPS can be improved by 1.9%，0.2%，and 9.4%，respectively，using only the dual-path adaptive feature calibration structure. When integrated with the GCM module，these improvements increase to 2.8%，0.64%，and 11.3%. The utilization of HVI color space decomposition yields in enhanced PSNR by 18.9% over the HSV baseline，SSIM by 6.4%，and perceptual quality by 64.3%. These three evaluation metrics improve by 4.9%，2.1%，and 36.4% compared with YCbCr. The use of only the HVI loss lacks pixel-space consistency constraints，resulting in large pixel errors and decreased performance across all three metrics，especially PSNR. By contrast，the use of only sRGB loss focuses on pixel-space enhancement but ignores the low-light probability distribution in the HVI color space，resulting in color imbalance. However，from a visualization perspective，the pixel-level errors and color distribution of using only HVI/sRGB loss functions are not intuitively observable.

The proposed LLIE method in this study effectively combines the advantages of a dual-branch structure and the HVI color space，achieving a balance between color fidelity and luminance artifact suppression in low-light image enhancement. GCDINet resolves the fundamental trade-off between color and artifacts through deterministic component decoupling and adaptive feature calibration，with ablation studies confirming the synergistic benefits of the GCFE and DAFC modules.