Image Clarity Research Articles

Evaluating Factors Shaping Real-Time Internet-of-Things-Based License Plate Recognition Using Single-Board Computer Technology

Reliable and cost-efficient license plate recognition (LPR) systems enhance security, traffic management, and automated toll collection in real-world applications. This study addresses optimal unique configurations for enhancing LPR system accuracy and reliability by evaluating the impact of camera angle, object velocity, and distance on the efficacy of real-time LPR systems. The Internet of Things (IoT) LPR framework is proposed and utilized on single-board computer (SBC) technology, such as the Raspberry Pi 4 platform, with a high-resolution webcam using advanced OpenCV and OCR–Tesseract algorithms applied. The research endeavors to simulate common deployment scenarios of the real-time LPR system and perform thorough testing by leveraging SBC computational capabilities and the webcam’s imaging capabilities. The testing process is not just comprehensive, but also meticulous, ensuring the system’s reliability in various operational settings. We performed extensive experiments with a hundred repetitions at diverse angles, velocities, and distances. An assessment of the data’s precision, recall, and F1 score indicates the accuracy with which Thai license plates are identified. The results show that camera angles close to 180° significantly reduce perspective distortion, thus enhancing precision. Lower vehicle speeds (<10 km/h) and shorter distances (<10 m) also improve recognition accuracy by reducing motion blur and improving image clarity. Images captured from shorter distances (approximately less than 10 m) are more accurate for high-resolution character recognition. This study substantially contributes to SBC technology utilizing IoT-based real-time LPR systems for practical, accurate, and cost-effective implementations.

A Systematic Literature Review: Performance Comparison of Edge Detection Operators in Medical Images

Medical images play a crucial role in the diagnosis of diseases. To make the diagnosis more accurate, the image should usually be enhanced first using image processing methods such as segmentation and edge detection stages. However, the complexity and noise that may arise in these images pose challenges in edge detection. Therefore, to portray the characteristics of edge detection operators, this research presents a systematic literature review of the performance of various edge detection operators in medical images, focusing on literature published between 2019 and 2023. After the selection process, 41 papers out of the initial 112 collected papers were chosen for further review. The study evaluates edge detection operators e.g., Canny, Sobel, Prewitt, Roberts, and Laplacian of Gaussian (LOG) on medical images such as X-rays, MRI, CT scans, ultrasound, Pap smears, and others. In the analysis, the accuracy, computational time, and response to noise of each operator are compared. The results indicate that despite longer computational times, Canny emerges as the most accurate operator, especially in Pap smear and CT scan images. The LOG operator offers high accuracy in MRI images with more efficient computational time. Evaluation of operator reliability against noise confirms the superiority of Canny. Furthermore, the future potential of edge detection in medical services is also reviewed. For instance, Canny, known for accurate and noise-resistant edges, enhances detection in complex CT-Scan and X-ray images. Meanwhile, LOG, handling artifacts with lower computational time, improves edge clarity in medical images. Potential applications include enhanced diagnosis, efficient patient monitoring, and improved image clarity in future medical services.

Deep learning for MRI lesion segmentation in rectal cancer.

Rectal cancer (RC) is a globally prevalent malignant tumor, presenting significant challenges in its management and treatment. Currently, magnetic resonance imaging (MRI) offers superior soft tissue contrast and radiation-free effects for RC patients, making it the most widely used and effective detection method. In early screening, radiologists rely on patients' medical radiology characteristics and their extensive clinical experience for diagnosis. However, diagnostic accuracy may be hindered by factors such as limited expertise, visual fatigue, and image clarity issues, resulting in misdiagnosis or missed diagnosis. Moreover, the distribution of surrounding organs in RC is extensive with some organs having similar shapes to the tumor but unclear boundaries; these complexities greatly impede doctors' ability to diagnose RC accurately. With recent advancements in artificial intelligence, machine learning techniques like deep learning (DL) have demonstrated immense potential and broad prospects in medical image analysis. The emergence of this approach has significantly enhanced research capabilities in medical image classification, detection, and segmentation fields with particular emphasis on medical image segmentation. This review aims to discuss the developmental process of DL segmentation algorithms along with their application progress in lesion segmentation from MRI images of RC to provide theoretical guidance and support for further advancements in this field.

Plant leaf disease recognition based on improved SinGAN and improved ResNet34.

The identification of plant leaf diseases is crucial in precision agriculture, playing a pivotal role in advancing the modernization of agriculture. Timely detection and diagnosis of leaf diseases for preventive measures significantly contribute to enhancing both the quantity and quality of agricultural products, thereby fostering the in-depth development of precision agriculture. However, despite the rapid development of research on plant leaf disease identification, it still faces challenges such as insufficient agricultural datasets and the problem of deep learning-based disease identification models having numerous training parameters and insufficient accuracy. This paper proposes a plant leaf disease identification method based on improved SinGAN and improved ResNet34 to address the aforementioned issues. Firstly, an improved SinGAN called Reconstruction-Based Single Image Generation Network (ReSinGN) is proposed for image enhancement. This network accelerates model training speed by using an autoencoder to replace the GAN in the SinGAN and incorporates a Convolutional Block Attention Module (CBAM) into the autoencoder to more accurately capture important features and structural information in the images. Random pixel Shuffling are introduced in ReSinGN to enable the model to learn richer data representations, further enhancing the quality of generated images. Secondly, an improved ResNet34 is proposed for plant leaf disease identification. This involves adding CBAM modules to the ResNet34 to alleviate the limitations of parameter sharing, replacing the ReLU activation function with LeakyReLU activation function to address the problem of neuron death, and utilizing transfer learning-based training methods to accelerate network training speed. This paper takes tomato leaf diseases as the experimental subject, and the experimental results demonstrate that: (1) ReSinGN generates high-quality images at least 44.6 times faster in training speed compared to SinGAN. (2) The Tenengrad score of images generated by the ReSinGN model is 67.3, which is improved by 30.2 compared to the SinGAN, resulting in clearer images. (3) ReSinGN model with random pixel Shuffling outperforms SinGAN in both image clarity and distortion, achieving the optimal balance between image clarity and distortion. (4) The improved ResNet34 achieved an average recognition accuracy, recognition precision, recognition accuracy (redundant as it's similar to precision), recall, and F1 score of 98.57, 96.57, 98.68, 97.7, and 98.17%, respectively, for tomato leaf disease identification. Compared to the original ResNet34, this represents enhancements of 3.65, 4.66, 0.88, 4.1, and 2.47%, respectively.

Immersive VR experience based on virtual robots in decorative material environment design simulation

Virtual robot and immersive VR experiences provide consumers with new ways of perception in the design process of decorative material environments. This experiential process is virtual, intuitive, and entertaining, and is widely recognized by consumers.The purpose of this study is to improve the processing effect of virtual images of decorative materials by image denoising algorithm. In this paper, the image acquisition and modeling technology based on sensor network and image denoising algorithm are used to process the acquired image data, and the processed image is applied to the virtual system of decorative materials. Through the application of image denoising algorithm, the noise in the image is reduced, and the clarity and authenticity of the virtual image of decorative materials are improved. This makes it possible to select the best decoration scheme more accurately when comprehensive calculation is carried out from various aspects such as aesthetic and functional data, and provides a new feasibility and method for environmental design and decoration material selection. The research results demonstrate that the system can achieve virtual image processing in environmental decoration scene design, thereby improving the level of indoor environmental art design.

Underwater image enhancement via color correction and multi-feature image fusion

Abstract The light attenuation underwater causes the actual underwater images to suffer from color cast, low contrast, and weak illumination. To address these issues, an effective fusion-based method is proposed, which realizes color correction, brightness adjustment, contrast, and detail enhancement of underwater images. Concretely, we first design an adaptive color correction method via dominant color channel judgment and lower color channel compensation. Then, we detect the brightness of each input image and propose a gamma correction function based on the gradient of the cumulative histogram to adjust the brightness of the low-light images. Subsequently, global histogram stretching and adaptive fractional differentiation techniques are employed to process the brightness-adjusted image, and then the global contrast-enhanced version and detail-enhanced version are generated respectively. To integrate the advantages of both versions, a channel fusion method based on the Lab color space is used to fuse the luminance and color of the two versions separately. The experimental results demonstrate the effectiveness of the proposed method in improving the color and illumination of underwater images, as well as enhancing the clarity of images. Moreover, the testing results on multiple datasets validate the excellent stability of this method.

Intelligent Detection of Marine Offshore Aquaculture with High-Resolution Optical Remote Sensing Images

The rapid and disordered expansion of artificial marine aquaculture areas has caused severe ecological and environmental problems. Accurate monitoring of offshore aquaculture areas is urgent and significant in order to support the scientific and sustainable management and protection of coastal marine resources. Artificial intelligence provides a valuable tool to improve marine resource monitoring. Deep learning methods have been widely used for marine object detection, but You Only Look Once (YOLO) models have not been employed for offshore aquaculture area monitoring. This study therefore evaluated the capacity of two well-known YOLO models, YOLOv5 and YOLOv7, to detect offshore aquaculture areas based on different high-resolution optical remote sensing imagery. Compared with YOLOv7 based on a satellite dataset, YOLOv5 increased the Precision value by approximately 3.29% (to 95.33%), Recall value by 3.02% (to 93.02%), mAP_0.5 by 2.03% (to 96.22%), and F1 score by 2.65% (to 94.16%). Based on the Google Earth dataset, YOLOv5 and YOLOv7 showed similar results. We found that the spatial resolution could affect the deep learning models’ performances. We used the Real-ESRGAN method to enhance the spatial resolution of satellite dataset and investigated whether super-resolution (SR) methods improved the detection accuracy of the YOLO models. The results indicated that despite improving the image clarity and resolution, the SR methods negatively affected the performance of the YOLO models for offshore aquaculture object detection. This suggests that attention should be paid to the use of SR methods before the application of deep learning models for object detection using remote sensing imagery.

The Impact of AI on Metal Artifacts in CBCT Oral Cavity Imaging.

This study aimed to assess the impact of artificial intelligence (AI)-driven noise reduction algorithms on metal artifacts and image quality parameters in cone-beam computed tomography (CBCT) images of the oral cavity. This retrospective study included 70 patients, 61 of whom were analyzed after excluding those with severe motion artifacts. CBCT scans, performed using a Hyperion X9 PRO 13 × 10 CBCT machine, included images with dental implants, amalgam fillings, orthodontic appliances, root canal fillings, and crowns. Images were processed with the ClariCT.AI deep learning model (DLM) for noise reduction. Objective image quality was assessed using metrics such as the differentiation between voxel values (ΔVVs), the artifact index (AIx), and the contrast-to-noise ratio (CNR). Subjective assessments were performed by two experienced readers, who rated overall image quality and artifact intensity on predefined scales. Compared with native images, DLM reconstructions significantly reduced the AIx and increased the CNR (p < 0.001), indicating improved image clarity and artifact reduction. Subjective assessments also favored DLM images, with higher ratings for overall image quality and lower artifact intensity (p < 0.001). However, the ΔVV values were similar between the native and DLM images, indicating that while the DLM reduced noise, it maintained the overall density distribution. Orthodontic appliances produced the most pronounced artifacts, while implants generated the least. AI-based noise reduction using ClariCT.AI significantly enhances CBCT image quality by reducing noise and metal artifacts, thereby improving diagnostic accuracy and treatment planning. Further research with larger, multicenter cohorts is recommended to validate these findings.

Contrast Enhancement Method Using Region-Based Dynamic Clipping Technique for LWIR-Based Thermal Camera of Night Vision Systems.

In the autonomous driving industry, there is a growing trend to employ long-wave infrared (LWIR)-based uncooled thermal-imaging cameras, capable of robustly collecting data even in extreme environments. Consequently, both industry and academia are actively researching contrast-enhancement techniques to improve the quality of LWIR-based thermal-imaging cameras. However, most research results only showcase experimental outcomes using mass-produced products that already incorporate contrast-enhancement techniques. Put differently, there is a lack of experimental data on contrast enhancement post-non-uniformity (NUC) and temperature compensation (TC) processes, which generate the images seen in the final products. To bridge this gap, we propose a histogram equalization (HE)-based contrast enhancement method that incorporates a region-based clipping technique. Furthermore, we present experimental results on the images obtained after applying NUC and TC processes. We simultaneously conducted visual and qualitative performance evaluations on images acquired after NUC and TC processes. In the visual evaluation, it was confirmed that the proposed method improves image clarity and contrast ratio compared to conventional HE-based methods, even in challenging driving scenarios such as tunnels. In the qualitative evaluation, the proposed method demonstrated upper-middle-class rankings in both image quality and processing speed metrics. Therefore, our proposed method proves to be effective for the essential contrast enhancement process in LWIR-based uncooled thermal-imaging cameras intended for autonomous driving platforms.

Improving Single-Image Super-Resolution with Dilated Attention

Single-image super-resolution (SISR) techniques have become a vital tool for improving image quality and clarity in the rapidly evolving field of digital imaging. Convolutional neural network (CNN) and transformer-based SISR techniques are very popular. However, CNN-based techniques are not suitable when capturing long-range dependencies, and transformer-based techniques suffer from computational complexity. To tackle these problems, this paper proposes a novel method called dilated attention-based single-image super-resolution (DAIR). It comprises three components: low-level feature extraction, multi-scale dilated transformer block (MDTB), and high-quality image reconstruction. A convolutional layer is used to extract the base features from low-resolution images, which lays the foundation for subsequent processing. Dilated attention is introduced to MDTB to enhance its ability to capture image features at different scales and ensure superior image details and structure recovery. After that, MDTB refines these features to extract multi-scale global attributes and effectively grasps images’ long-distance relationships and features across multiple scales. Finally, low-level features obtained from feature extraction and multi-scale global features obtained from MDTB are aggregated to reconstruct high-resolution images. The comparison with existing methods validates the efficacy of the proposed method and demonstrates its advantage in improving image resolution and quality.

Hyperspectral-multispectral image fusion using subspace decomposition and Elastic Net Regularization

ABSTRACT The fusion of hyperspectral and multispectral images presents a challenge as it involves blending a low-resolution hyperspectral image (HSI) with a corresponding multispectral image (MSI) to produce a high-resolution hyperspectral image (HRI). A number of existing techniques have limitations; for instance, matrix decomposition-based approaches fail to retain adequate spatial and spectral image information during fusion, while tensor decomposition-based processes have high computational overhead. In this paper, we propose a novel method for fusing hyperspectral and multispectral images. Our method leverages the strong correlation among the spectral bands of hyperspectral images and employs the SVD technique to extract spectral feature subspaces. This approach results in a more compact and representative feature space for fusion. Secondly, the proposed method utilizes Elastic Net regularization in combination with L 1 and L 2 regularization for effective feature selection of highly covariate features. Weighted group sparse regularization is employed to enhance the fusion effect, enabling better representation of the image’s structure and features. The algorithm is subsequently evaluated on multiple datasets to confirm its effectiveness. The results of the experiment suggest that the suggested algorithm greatly enhances the spatial resolution and visual clarity of HSI images while preserving the spectral characteristics when compared to conventional methods for fusion of hyperspectral images. Additionally, the constraint for regulation can competently reduce any noise or artefacts, thereby boosting image discrimination. To summarize, the utilization of a sparse tensor-based hyperspectral image fusion algorithm with subspace learning provides an efficacious approach for processing hyperspectral imagery. This method is capable of improving spatial resolution, extracting advantageous features from hyperspectral imagery, and ultimately supports the process of remote sensing image analysis and application.

Advancements in Edge Detection Techniques for Image Enhancement: A Comprehensive Review

Edge detection is a fundamental algorithm in image processing and computer vision, widely applied in various domains such as medical imaging and autonomous driving. This comprehensive literature review critically evaluates the latest edge detection methods, encompassing classical approaches (Sobel, Canny, and Prewitt) and advanced techniques based on deep learning, fuzzy logic, and optimization algorithms. The review summarises the significant contributions and advancements in the field by synthesizing insights from numerous research papers. It also examines the combination of edge detection with current image processing methods and discusses its impact on real-life applications. The review highlights the strengths and limitations of existing edge detection strategies and proposes future avenues for investigation. Various research shows that classical edge detection methods like Sobel, Canny, and Prewitt still play a significant role in the field. However, advanced methods utilizing deep learning, fuzzy logic, and optimization algorithms have shown promising results in enhancing edge detection accuracy. Combining edge detection with current image processing methods has demonstrated improved clarity and interpretation of images in real-life applications, including medical imaging and machine learning systems. Despite the progress made, there are still limitations and challenges in existing edge detection strategies that require further investigation. Future research should address these shortcomings and explore new edge detection algorithm development avenues. By understanding the current state of the art and its implications, researchers and practitioners can make informed decisions and contribute to advancing edge detection in image processing and analysis. Overall, this review serves as a valuable guide for researchers and practitioners working in the field, providing a thorough understanding of the state-of-the-art edge detection techniques, their implications for image processing, and their potential for further development.

Quantitative Evaluation of Color Enhancement Methods for Underwater Photogrammetry in Very Shallow Water: a Case Study

Abstract. Underwater photogrammetry is often hampered by chromatic aberration, leading to degraded 2D and 3D products. This study investigates the effectiveness of various color enhancement methods in addressing these challenges.Theoretical considerations indicate that light penetration depth varies inversely with wavelength, causing underwater images to exhibit a blue or green cast with increasing depth. Color enhancement techniques can restore natural colors by compensating for this spectral attenuation. Additionally, scattering, caused by light reflected by particles in the water, can introduce haze into underwater images. Color enhancement can mitigate scatter and improve image clarity. In this contribution, to quantitatively evaluate color enhancement methods, we compare original images with images processed using gray-world assumption methods and physical methods that account for the physical properties of light underwater. Using artificial intelligence (AI) for underwater image color enhancement, a data-driven approach was also employed. These methods were applied to a case study concerning a Roman Navis Lapidaria shipwreck carrying five monumental cipollino marble columns at a depth of 4.5 meters in the Porto Cesareo Marine Protected Area (Italy). These methods were compared quantitatively and qualitatively, and the results are presented and discussed.

DehazeModel: Enhancing Image Clarity with an Encoder-Decoder CNN Approach

We introduce DehazeModel, a Convolutional Neural Network (CNN) tailored for the purpose of enhancing single images by removing haze. DehazeModel comprises several components, including pre-processing, image dehazing, and post-processing modules. The pre-processing module, which is trainable, generates enhanced inputs featuring a wider array of characteristics compared to manually chosen pre-processing methods. The Image Dehazing module employs a novel encoder-decoder framework, overcoming common issues found in traditional multi-scale methods. Moreover, the post-processing module aids in minimizing artifacts in the resultant output. Experimental findings indicate that DehazeModel surpasses existing state-of-the-art techniques on the RESIDE dataset. Our experiments, conducted on both indoor and outdoor images, illustrate the robustness of our approach across various scenarios, independent of atmospheric scattering effects.

Detection and Quantification of Arnica montana L. Inflorescences in Grassland Ecosystems Using Convolutional Neural Networks and Drone-Based Remote Sensing

Arnica montana L. is a medicinal plant with significant conservation importance. It is crucial to monitor this species, ensuring its sustainable harvesting and management. The aim of this study is to develop a practical system that can effectively detect A. montana inflorescences utilizing unmanned aerial vehicles (UAVs) with RGB sensors (red–green–blue, visible light) to improve the monitoring of A. montana habitats during the harvest season. From a methodological point of view, a model was developed based on a convolutional neural network (CNN) ResNet101 architecture. The trained model offers quantitative and qualitative assessments of A. montana inflorescences detected in semi-natural grasslands using low-resolution imagery, with a correctable error rate. The developed prototype is applicable in monitoring a larger area in a short time by flying at a higher altitude, implicitly capturing lower-resolution images. Despite the challenges posed by shadow effects, fluctuating ground sampling distance (GSD), and overlapping vegetation, this approach revealed encouraging outcomes, particularly when the GSD value was less than 0.45 cm. This research highlights the importance of low-resolution image clarity, on the training data by the phenophase, and of the need for training across different photoperiods to enhance model flexibility. This innovative approach provides guidelines for mission planning in support of reaching sustainable management goals. The robustness of the model can be attributed to the fact that it has been trained with real-world imagery of semi-natural grassland, making it practical for fieldwork with accessible portable devices. This study confirms the potential of ResNet CNN models to transfer learning to new plant communities, contributing to the broader effort of using high-resolution RGB sensors, UAVs, and machine-learning technologies for sustainable management and biodiversity conservation.

Mixed-mode oscillations of an atomic force microscope in tapping mode.

In the phenomenon of mixed-mode oscillations, transitions between large-amplitude and small-amplitude oscillations may lead to anomalous jitter in the probe of a tapping mode atomic force microscope (TM-AFM) during the scanning process, thereby affecting the accuracy and clarity of the topographical images of the tested sample's surface. This work delves deeply into various mixed-mode oscillations and the corresponding formation mechanisms in TM-AFM under low-frequency resonant excitation. Through a detailed analysis of bifurcation sets of the fast subsystem, we found that the system's mixed-mode oscillations encompass the typical two coexisting branches and the novel three coexisting branches of equilibrium point attractors. In the stable case, a certain transition pattern in phase trajectory can be observed involving two jumps and four jumps, switching between quiescent and spiking states. In the bi-stable case, the trajectory undergoes distinct transitions decided by whether to pass through or crossover the middle branch of attractors when bifurcation occurs. By applying basin of attraction and fast-slow analysis methods, we unfold the dynamic mechanism of mixed-mode oscillations with distinct switching patterns. Our research contributes to a better understanding of complex oscillations of TM-AFM and provides valuable insights for improving image quality and measurement precision while mitigating detrimental oscillations.

Preparation of Retinal Samples for Volume Electron Microscopy.

Volume electron microscopy (Volume EM) has emerged as a powerful tool for visualizing the 3D structure of cells and tissues with nanometer-level precision. Within the retina, various types of neurons establish synaptic connections in the inner and outer plexiform layers. While conventional EM techniques have yielded valuable insights into retinal subcellular organelles, their limitation lies in providing 2D image data, which can hinder accurate measurements. For instance, quantifying the size of three distinct synaptic vesicle pools, crucial for synaptic transmission, is challenging in 2D. Volume EM offers a solution by providing large-scale, high-resolution 3D data. It is worth noting that sample preparation is a critical step in Volume EM, significantly impacting image clarity and contrast. In this context, we outline a sample preparation protocol for the 3D reconstruction of photoreceptor axon terminals in the retina. This protocol includes three key steps: retina dissection and fixation, sample embedding processes, and selection of the area of interest.

Vehicle detection algorithm for foggy based on improved AOD-Net

To strengthen the safety monitoring of foggy road traffic and maintain the safety of vehicle driving on foggy roads, image dehazing algorithms are used to improve the clarity of road images detected in foggy environments, thereby improving the detection ability and monitoring efficiency of intelligent transportation systems for vehicle targets. Due to the low accuracy of vehicle detection and serious problem of missed detections in haze environments, this paper proposes an improved All-in-One Dehazing Network (AOD-Net) algorithm for detecting foggy vehicles, which adds batch normalization (BN) layers after each layer of convolution in AOD-Net, accelerating the convergence of the model and controlling overfitting. To enhance image detail information, an effective pyramid-shaped PSA attention module is embedded to extract richer feature information, enrich model representation, and improve the loss function to a multi-scale structural similarity (MS-SSIM) + L1 mixed loss function, thereby improving the quality, brightness, and contrast of dehazing images. Compared with current image dehazing algorithms, the dehazing quality of our algorithm is superior to other dehazing algorithms, such as dark channel prior (DCP), Dehaze-Net, and Fusion Feature Attention Network (FFA-Net). Compared with AOD-Net, the improved algorithm has increased the peak signal-to-noise ratio by 3.23 dB. At the same time, after the improved AOD-Net image dehazing processing, YOLOv7 object detection was performed and experimentally validated on a real foggy dataset. The results showed that compared with the previous method, it had better recognition performance in foggy detection and recognition, and higher detection accuracy for vehicles.

Image recovery method for underwater targets with complex polarization characteristics

Polarization imaging techniques have been effective in improving the clarity of turbid underwater images affected by water scattering. These techniques offer valuable additional information compared to traditional methods. However, previous descattering methods have mostly been designed for targets with uniform distribution of polarimetric characteristics. Therefore, targets with complex polarization characteristics have non-uniform distribution of polarization characteristics when dealing with such problems, additional parameter estimation errors can limit the results of image recovery. This paper proposes what we believe is a novel approach to address this issue. The method involves obtaining a new set of images using the polarization images vector space transformation method. The angle of polarization (AOP) of the target reflected light is estimated using the additivity law of the Stokes vector. This information is then combined with the physical model of underwater imaging to recover the underwater images affected by scattering. Experimental results conducted under varying levels of water turbidity validate the effectiveness of the proposed method. Moreover, the proposed method significantly broadens the range of application scenarios.

A Novel Lightweight Model for Underwater Image Enhancement.

Underwater images suffer from low contrast and color distortion. In order to improve the quality of underwater images and reduce storage and computational resources, this paper proposes a lightweight model Rep-UWnet to enhance underwater images. The model consists of a fully connected convolutional network and three densely connected RepConv blocks in series, with the input images connected to the output of each block with a Skip connection. First, the original underwater image is subjected to feature extraction by the SimSPPF module and is processed through feature summation with the original one to be produced as the input image. Then, the first convolutional layer with a kernel size of 3 × 3, generates 64 feature maps, and the multi-scale hybrid convolutional attention module enhances the useful features by reweighting the features of different channels. Second, three RepConv blocks are connected to reduce the number of parameters in extracting features and increase the test speed. Finally, a convolutional layer with 3 kernels generates enhanced underwater images. Our method reduces the number of parameters from 2.7 M to 0.45 M (around 83% reduction) but outperforms state-of-the-art algorithms by extensive experiments. Furthermore, we demonstrate our Rep-UWnet effectively improves high-level vision tasks like edge detection and single image depth estimation. This method not only surpasses the contrast method in objective quality, but also significantly improves the contrast, colorimetry, and clarity of underwater images in subjective quality.