Highly Differentiated Target Detection under Extremely Low-Light Conditions Based on Improved YOLOX Model

This paper expounds upon a novel target detection methodology distinguished by its elevated discriminatory efficacy,specifically tailored for environments characterized by markedly low luminance levels.Conventional methodologies struggle with the challenges posed by luminosity fluctuations,especially in settings characterized by diminished radiance,further exacerbated by the utilization of suboptimal imaging instrumentation.The envisioned approach mandates a departure from the conventional YOLOX model,which exhibits inadequacies in mitigating these challenges.To enhance the efficacy of this approach in low-light conditions,the dehazing algorithm undergoes refinement,effecting a discerning regulation of the transmission rate at the pixel level,reducing it to values below 0.5,thereby resulting in an augmentation of image contrast.Subsequently,the coiflet wavelet transform is employed to discern and isolate high-discriminatory attributes by dismantling low-frequency image attributes and extracting high-frequency attributes across divergent axes.The utilization of CycleGAN serves to elevate the features of low-light imagery across an array of stylistic variances.Advanced computational methodologies are then employed to amalgamate and conflate intricate attributes originating from images characterized by distinct stylistic orientations,thereby augmenting the model’s erudition potential.Empirical validation conducted on the PASCAL VOC and MS COCO 2017 datasets substantiates pronounced advancements.The refined low-light enhancement algorithm yields a discernible 5.9%augmentation in the target detection evaluation index when compared to the original imagery.Mean Average Precision(mAP)undergoes enhancements of 9.45%and 0.052%in low-light visual renditions relative to conventional YOLOX outcomes.The envisaged approach presents a myriad of advantages over prevailing benchmark methodologies in the realm of target detection within environments marked by an acute scarcity of luminosity.

More Than Lightening:A Self-Supervised Low-Light Image Enhancement Method Capable for Multiple Degradations

Low-light images suffer from low quality due to poor lighting conditions,noise pollution,and improper settings of cameras.To enhance low-light images,most existing methods rely on normal-light images for guidance but the collection of suitable normal-light images is difficult.In contrast,a self-supervised method breaks free from the reliance on normal-light data,resulting in more convenience and better generalization.Existing self-supervised methods primarily focus on illumination adjustment and design pixel-based adjustment methods,resulting in remnants of other degradations,uneven brightness and artifacts.In response,this paper proposes a self-supervised enhancement method,termed as SLIE.It can handle multiple degradations including illumination attenuation,noise pollution,and color shift,all in a self-supervised manner.Illumination attenuation is estimated based on physical principles and local neighborhood information.The removal and correction of noise and color shift removal are solely realized with noisy images and images with color shifts.Finally,the comprehensive and fully self-supervised approach can achieve better adaptability and generalization.It is applicable to various low light conditions,and can reproduce the original color of scenes in natural light.Extensive experiments conducted on four public datasets demonstrate the superiority of SLIE to thirteen state-of-the-art methods.Our code is available at https://github.com/hanna-xu/SLIE.

Localized electric-field-enhanced low-light detection by a 2D SnS visible-light photodetector

Due to their excellent carrier mobility,high absorption coefficient and narrow bandgap,most 2D IVA metal chalcogenide semiconductors(GIVMCs,metal=Ge,Sn,Pb;chalcogen=S,Se)are regarded as promising candidates for realizing high-performance photodetectors.We synthesized high-quality two-dimensional(2D)tin sulfide(SnS)nanosheets using the physical vapor deposition(PVD)method and fabricated a 2D SnS visible-light photodetector.The photodetector exhibits a high photoresponsivity of 161 A·W-1 and possesses an external quantum efficiency of 4.45×10^(4)%,as well as a detectivity of 1.15×10^(9) Jones under 450 nm blue light illumination.Moreover,under poor illumination at optical densities down to 2 mW·cm^(-2),the responsivity of the device is higher than that at stronger optical densities.We suggest that a photogating effect in the 2D SnS photodetector is mainly responsible for its low-light responsivity.Defects and impurities in 2D SnS can trap carriers and form localized electric fields,which can delay the recombination process of electron-hole pairs,prolong carrier lifetimes,and thus improve the low-light responsivity.This work provides design strategies for detecting low levels of light using photodetectors made of 2D materials.

RF-Net: Unsupervised Low-Light Image Enhancement Based on Retinex and Exposure Fusion

Low-light image enhancement methods have limitations in addressing issues such as color distortion,lack of vibrancy,and uneven light distribution and often require paired training data.To address these issues,we propose a two-stage unsupervised low-light image enhancement algorithm called Retinex and Exposure Fusion Network(RFNet),which can overcome the problems of over-enhancement of the high dynamic range and under-enhancement of the low dynamic range in existing enhancement algorithms.This algorithm can better manage the challenges brought about by complex environments in real-world scenarios by training with unpaired low-light images and regular-light images.In the first stage,we design a multi-scale feature extraction module based on Retinex theory,capable of extracting details and structural information at different scales to generate high-quality illumination and reflection images.In the second stage,an exposure image generator is designed through the camera response mechanism function to acquire exposure images containing more dark features,and the generated images are fused with the original input images to complete the low-light image enhancement.Experiments show the effectiveness and rationality of each module designed in this paper.And the method reconstructs the details of contrast and color distribution,outperforms the current state-of-the-art methods in both qualitative and quantitative metrics,and shows excellent performance in the real world.

Low-Light Enhancer for UAV Night Tracking Based on Zero-DCE++

Unmanned aerial vehicle (UAV) target tracking tasks can currently be successfully completed in daytime situations with enough lighting, but they are unable to do so in nighttime scenes with inadequate lighting, poor contrast, and low signal-to-noise ratio. This letter presents an enhanced low-light enhancer for UAV nighttime tracking based on Zero-DCE++ due to its ad-vantages of low processing cost and quick inference. We developed a light-weight UCBAM capable of integrating channel information and spatial features and offered a fully considered curve projection model in light of the low signal-to-noise ratio of night scenes. This method significantly improved the tracking performance of the UAV tracker in night situations when tested on the public UAVDark135 and compared to other cutting-edge low-light enhancers. By applying our work to different trackers, this search shows how broadly applicable it is.

Novel Frame Shift and Integral Technique for Enhancing Low-Light-Level Moving Images

A novel frame shift and integral technique for the enhancement of low light level moving image sequence is introduced. According to the technique, motion parameters of target are measured by algorithm based on difference processing. To obtain spatial relativity, images are shifted according to the motion parameters. As a result, the processing of integral and average can be applied to images that have been shifted. The technique of frame shift and integral that includes the algorithm of motion parameter determination is discussed, experiments with low light level moving image sequences are also described. The experiment results show the effectiveness and the robustness of the parameter determination algorithm, and the improvement in the signal-to-noise ratio （SNR） of low light level moving images.

MAGAN:Unsupervised Low-Light Image Enhancement Guided by Mixed-Attention

Most learning-based low-light image enhancement methods typically suffer from two problems.First,they require a large amount of paired data for training,which are difficult to acquire in most cases.Second,in the process of enhancement,image noise is difficult to be removed and may even be amplified.In other words,performing denoising and illumination enhancement at the same time is difficult.As an alternative to supervised learning strategies that use a large amount of paired data,as presented in previous work,this paper presents an mixed-attention guided generative adversarial network called MAGAN for low-light image enhancement in a fully unsupervised fashion.We introduce a mixed-attention module layer,which can model the relationship between each pixel and feature of the image.In this way,our network can enhance a low-light image and remove its noise simultaneously.In addition,we conduct extensive experiments on paired and no-reference datasets to show the superiority of our method in enhancing low-light images.

DEANet: Decomposition Enhancement and Adjustment Network for Low-Light Image Enhancement

Poor illumination greatly affects the quality of obtained images.In this paper,a novel convolutional neural network named DEANet is proposed on the basis of Retinex for low-light image enhancement.DEANet combines the frequency and content information of images and is divided into three subnetworks:decomposition,enhancement,and adjustment networks,which perform image decomposition;denoising,contrast enhancement,and detail preservation;and image adjustment and generation,respectively.The model is trained on the public LOL dataset,and the experimental results show that it outperforms the existing state-of-the-art methods regarding visual effects and image quality.

Retinex based low-light image enhancement using guided filtering and variational framework

A new image enhancement algorithm based on Retinex theory is proposed to solve the problem of bad visual effect of an image in low-light conditions. First, an image is converted from the RGB color space to the HSV color space to get the V channel. Next, the illuminations are respectively estimated by the guided filtering and the variational framework on the V channel and combined into a new illumination by average gradient. The new reflectance is calculated using V channel and the new illumination. Then a new V channel obtained by multiplying the new illumination and reflectance is processed with contrast limited adaptive histogram equalization(CLAHE). Finally, the new image in HSV space is converted back to RGB space to obtain the enhanced image. Experimental results show that the proposed method has better subjective quality and objective quality than existing methods.

Low-light color image enhancement based on NSST

In order to improve the visibility and contrast of low-light images and better preserve the edge and details of images,a new low-light color image enhancement algorithm is proposed in this paper.The steps of the proposed algorithm are described as follows.First,the image is converted from the red,green and blue(RGB)color space to the hue,saturation and value(HSV)color space,and the histogram equalization(HE)is performed on the value component.Next,non-subsampled shearlet transform(NSST)is used on the value component to decompose the image into a low frequency sub-band and several high frequency sub-bands.Then,the low frequency sub-band and high frequency sub-bands are enhanced respectively by Gamma correction and improved guided image filtering(IGIF),and the enhanced value component is formed by inverse NSST transform.Finally,the image is converted back to the RGB color space to obtain the enhanced image.Experimental results show that the proposed method not only significantly improves the visibility and contrast,but also better preserves the edge and details of images.

Filter-cluster attention based recursive network for low-light enhancement

The poor quality of images recorded in low-light environments affects their further applications.To improve the visibility of low-light images,we propose a recurrent network based on filter-cluster attention(FCA),the main body of which consists of three units:difference concern,gate recurrent,and iterative residual.The network performs multi-stage recursive learning on low-light images,and then extracts deeper feature information.To compute more accurate dependence,we design a novel FCA that focuses on the saliency of feature channels.FCA and self-attention are used to highlight the low-light regions and important channels of the feature.We also design a dense connection pyramid(DenCP)to extract the color features of the low-light inversion image,to compensate for the loss of the image's color information.Experimental results on six public datasets show that our method has outstanding performance in subjective and quantitative comparisons.

Toward Robust and Efficient Low-Light Image Enhancement:Progressive Attentive Retinex Architecture Search

In recent years,learning-based low-light image enhancement methods have shown excellent performance,but the heuristic design adopted by most methods requires high engineering skills for developers,causing expensive inference costs that are unfriendly to the hardware platform.To handle this issue,we propose to automatically discover an efficient architecture,called progressive attentive Retinex network(PAR-Net).We define a new attentive Retinex framework by introducing the attention mechanism to strengthen structural representation.A multi-level search space containing micro-level on the operation and macro-level on the cell is established to realize meticulous construction.To endow the searched architecture with the hardware-aware property,we develop a latency-constrained progressive search strategy that successfully improves the model capability by explicitly expressing the intrinsic relationship between different models defined in the attentive Retinex framework.Extensive quantitative and qualitative experimental results fully justify the superiority of our proposed approach against other state-of-the-art methods.A series of analytical evaluations is performed to illustrate the validity of our proposed algorithm.

6D pose annotation and pose estimation method for weak-corner objects under low-light conditions

In unstructured environments such as disaster sites and mine tunnels,it is a challenge for robots to estimate the poses of objects under complex lighting backgrounds,which limit their operation.Owing to the shadows produced by a point light source,the brightness of the operation scene is seriously unbalanced,and it is difficult to accurately extract the features of objects.It is especially difficult to accurately label the poses of objects with weak corners and textures.This study proposes an automatic pose annotation method for such objects,which combine 3D-2D matching projection and rendering technology to improve the efficiency of dataset annotation.A 6D object pose estimation method under low-light conditions(LP_TGC)is then proposed,including(1)a light preprocessing neural network model based on a low-light preprocessing module(LPM)to balance the brightness of a picture and improve its quality;and(2)a 6D pose estimation model(TGC)based on the keypoint matching.Four typical datasets are constructed to verify our method,the experimental results validated and demonstrated the effectiveness of the proposed LP_TGC method.The estimation model based on the preprocessed image can accurately estimate the pose of the object in the mentioned unstructured environments,and it can improve the accuracy by an average of~3%based on the ADD metric.

基于双曲嵌入的露天矿区暗光环境下道路多目标检测模型

露天矿环境特殊,道路场景复杂多变,在光照不足时会导致矿区道路多目标识别不清、定位不准,进而影响检测效果,给矿区无人矿用卡车的安全行驶带来严重安全隐患。目前的道路障碍物检测模型不能有效解决矿区暗光环境对模型检测效果的影响,同时对矿区小目标障碍物的识别也有较大误差,不适用于矿区特殊环境下障碍物的检测与识别。针对上述问题,提出了一种基于双曲嵌入的露天矿区暗光环境下多目标检测模型。首先,在模型的图像预处理阶段引入卷积神经网路Retinex-Net对暗图像进行增强,提高图像清晰度;然后,针对数据集中特征过多而无重点偏好的问题,在加强特征提取部分添加全局注意力机制,聚集3个维度上更关键的特征信息;最后,在检测模型预测阶段引入双曲全连接层,以减少特征丢失,并防止过拟合现象。实验结果表明:(1)基于双曲嵌入的露天矿区暗光环境下道路多目标检测模型不仅对露天矿区暗光环境下的大尺度目标具有较高的分类与定位精度,对矿用卡车及较远距离的小尺度目标即行人也可准确检测与定位,能够满足无人矿用卡车在矿区特殊环境下驾驶的安全需求。(2)模型的检测准确率达98.6%,检测速度为51.52帧/s,较SSD、YOLOv4、YOLOv5、YOLOx、YOLOv7分别提高20.31%,18.51%,10.53%,8.39%,13.24%,对于矿区道路上的行人、矿用卡车及挖机的检测精度达97%以上。

低温弱光处理对茄子不同时期花青素含量及果实品质的影响

【目的】探究低温、弱光、低温弱光处理对茄子幼苗期、花期、果期花青素含量的影响,以及对茄子品质的影响,为茄子的优质培育以及高产栽培奠定理论基础。【方法】以紫黑茄秀娘为试验材料,分别在幼苗期、花期、果期进行低温(18℃/13℃,250μmol·m^(-2)·s^(-1))、弱光(25℃/20℃,120μmol·m^(-2)·s^(-1))、低温弱光(18℃/13℃,120μmol·m^(-2)·s^(-1))、CK(25℃/20℃,250μmol·m^(-2)·s^(-1))等4个处理,测定幼苗期形态及生理特性,不同时期、不同部位的花青素,以及果期果实的品质。【结果】低温弱光胁迫对幼苗生长存在显著影响,在幼苗期低温对幼苗生长及生理影响显著大于弱光及低温弱光,花青素含量均表现为根<叶片<叶脉<茎;在花期,花青素含量依次为花萼<花瓣;在果期,花青素含量依次为果肉<果柄<果皮。茄子不同时期受到胁迫后,不同部位的花青素含量均呈现弱光<CK<低温弱光<低温,各胁迫下果实色泽指数依次为弱光<CK<低温弱光<低温,可溶性糖含量、可溶性蛋白含量、类黄酮含量、总酚含量均呈现低温<低温弱光<弱光<CK。【结论】低温促进花青素合成;弱光抑制花青素合成;在低温弱光双因素互作下,低温因素对花青素含量的影响起主导作用,花青素的合成大于降解,花青素含量增加。低温、弱光、低温弱光胁迫下茄子品质均下降,其中,低温胁迫对茄子的品质影响最大。

基于弱光增强与YOLO算法的锯链缺陷检测方法

在基于机器视觉的锯链缺陷实时检测过程中,油污、粉尘等因素影响图像亮度和质量,导致目标检测网络的特征提取能力下降。为保证复杂环境下锯链缺陷检测的准确率,本文设计了一种结合弱光增强和YOLOv3算法的锯链自动化缺陷检测方法。首先使用RRDNet网络自适应增强锯链图像亮度,恢复图像暗区的细节特征;然后采用改进YOLOv3算法对锯链零件进行缺陷检测,增加FPN结构特征输出图层,利用K-means聚类算法对先验框参数重新聚类,并引入GIoU损失函数来提高小目标的缺陷检测精度。最后搭建一套锯链缺陷在线检测系统,对所提方法进行验证。实验结果表明,该方法能够显著提高弱光环境下的锯链图像照度、恢复图像细节,改进YOLOv3算法的mAP值为92.88%,相比原始YOLOv3提高14%,最终系统整体的漏检率降低到3.2%,过检率也降低到9.1%。所提出的方法可实现弱光场景下锯链缺陷的在线检测,并且对多种缺陷有着较高的检测精度。

一种面向机器视觉感知的暗光图像增强网络

低光照等恶劣环境下的目标检测一直都是难点,低光照和多雾因素往往会导致图像出现可视度低、噪声大等情况,严重干扰目标检测的检测精度。针对上述问题,提出了一个面向机器视觉感知的低光图像增强网络MVP-Net,并与YOLOv3目标检测网络整合,构建了端到端的增强检测框架MVP-YOLO。MVP-Net采用了逆映射网络技术,将常规RGB图像转换为伪RAW图像特征空间,并提出了伪ISP增强网络DOISP进行图像增强。MVP-Net旨在发挥RAW图像在目标检测中的潜在优势,同时克服其在直接应用时所面临的限制。模型在多个真实场景暗光数据上取得了优于先前工作效果并且能够适应多种不同架构的检测器。其端到端检测框mAP(50%)指标达到了78.3%,比YOLO检测器提高了1.85%。

低光照下的无人机异物检测与定位

为解决无人机在低光照环境下的巡检过程中,不能对场景中的异物进行识别与定位,导致后续智能算法无法获得环境语义信息的问题.本文提出一种将ORB-SLAM2算法与适用于低光照目标检测改进的YOLOv5模型进行信息融合的方法.首先,通过RGB-D相机自采集低光照数据集进行深度学习训练及融合算法验证.然后,结合关键帧信息、目标检测模块的输出结果以及相机的固有信息完成目标像素坐标提取.最后,通过关键帧信息和像素坐标完成目标物体相对世界坐标系的位置解算.本文实现了低光照环境下目标物体较为准确的识别和目标物体在世界坐标系中分米级的定位,为低光照环境下无人机智能巡检提供了一种有效的解决方案.

基于HSV空间的煤矿井下低光照图像增强方法

针对煤矿井下采集到的图像对比度低、光照不均和细节信息弱等问题,提出一种基于色相-饱和度-明度(Hue-Saturation-Value,HSV)颜色空间的煤矿井下低光照图像增强方法。该方法基于图像的HSV空间,通过对低光照图像的亮度通道V通道的主要结构和边缘细节分别进行对比度增强,这样可以更好地抑制图像细节丢失,同时可以较好地再现原图中的轮廓和纹理细节。首先,将输入的煤矿井下低光照图像转换到HSV空间,利用相对全变分滤波(RTV)与改进的边窗滤波(SWF),分别对提取的V通道图像进行主要结构提取和轮廓边缘保留,对其非线性灰度拉伸后利用主成分分析融合技术(PCA)重构V通道图像,即融合V通道图像的主要结构和精细结构,最后合成图像,完成图像增强。通过实验验证,提出的基于HSV空间的煤矿井下低光照图像增强方法,在色彩和边缘模糊处理等方面表现良好,在煤矿井下工作面等环境中,对图像进行定量和定性实验,结果表明,与6种方法相比,增强图像的对比度、自然度和图像细节方面表现更好。

从局部到全局的零参考低照度图像增强方法

为解决现有的低照度图像增强方法存在的色彩失真、细节损失以及暗区增强不足和亮区增强过度导致低照度图像增强效果不理想的问题,提出了一种从局部到全局的零参考低照度图像增强方法。采用局部照度增强对低照度图像进行像素级增强,改进了自适应光照映射估计函数,提升了照度调整能力,避免了生成大量的迭代参数,提高了模型的推理速度;采用基于Transformer结构的全局图像调整对局部增强后的图像进行全局调整,解决了亮区照度增强过度的曝光问题和暗区照度增强不足的问题,提升了图像的整体对比度;优化损失函数,对低照度图像特征和增强图像特征进行相似性约束,提升了目标检测精度。实验结果表明,LOL数据集上的客观指标峰值信噪比和结构相似性达到了20.18 dB和0.80,MIT-Adobe FiveK数据集上达到了23.31 dB和0.87,ExDark数据集上增强后图像的目标检测精度提高了7.6%,有效提升了低照度图像可视化质量和目标检测效果。