融入混合注意力的低缩放因子SeamCarving篡改检测算法

针对现有的Seam Carving篡改检测算法对于低缩放因子情况存在检测精度不高、鲁棒性不强的问题,提出一种融入混合注意力机制的Seam Carving篡改检测算法。首先,利用BayarConv2D约束卷积对图像进行预处理,充分学习图像的噪声特征,并通过矩阵乘法与RGB图像进行特征融合;然后,采用ResNet作为骨干网络进行特征学习,引入残差传播和残差反馈机制,凸显Seam Carving的操作痕迹;最后,利用混合注意力机制同时提取相邻位置和通道之间的特征,更好地捕捉全局特征,进而将其输入全连接层进行分类。实验结果表明,在BOSSbase1.01数据集上,当缩放因子为1%和9%时,检测精度分别达到了89.48%和97.94%,优于现有主流方法,同时具有较低的计算复杂度和较好的鲁棒性,能够抵抗JPEG压缩攻击。

基于动态规划的Seam Carving裁剪算法

随着科技的发展,图像应用和分享越来越普遍。图像尺寸调整在不同设备和屏幕上的显示非常重要。传统的方法只是简单地复制或计算像素值,而不考虑图像内容的重要性。然而,基于动态规划的Seam Carving裁剪算法提出了一种更优化的方法。该算法采用多尺度处理,对不同尺度的图像进行Seam Carving操作,并将处理结果整合起来,以实现更全面、更精细的图像调整。

From mice to humans:a need for comparable results in mammalian neuroplasticity

Brain plasticity-A universal tool with many variations:The study of brain plasticity has been gaining interest since almost a century and has now reached a huge amount of information(>80,000 results in PubMed).Overall,different types of plasticity,including stem cell-driven genesis of new neurons(adult neurogenesis),cells in arrested maturation(dormant neurons),neuro-glial and synaptic plasticity,can coexist and contribute to grant plastic changes in the brain,from a cellular to system level(Benedetti and Couillard-Despres,2022;Bonfanti et al.,2023).

Contribution of mechanical forces to structural synaptic plasticity:insights from 3D cellular motility mechanisms

Cells,tissues,and organs are constantly subjected to the action of mechanical forces from the extracellular environment-and the nervous system is no exception.Cell-intrinsic properties such as membrane lipid composition,abundance of mechanosensors,and cytoskeletal dynamics make cells more or less likely to sense these forces.Intrinsic and extrinsic cues are integrated by cells and this combined information determines the rate and dynamics of membrane protrusion growth or retraction(Yamada and Sixt,2019).Cell protrusions are extensions of the plasma membrane that play crucial roles in diverse contexts such as cell migration and neuronal synapse formation.In the nervous system,neurons are highly dynamic cells that can change the size and number of their pre-and postsynaptic elements(called synaptic boutons and dendritic spines,respectively),in response to changes in the levels of synaptic activity through a process called plasticity.Synaptic plasticity is a hallmark of the nervous system and is present throughout our lives,being required for functions like memory formation or the learning of new motor skills(Minegishi et al.,2023;Pillai and Franze,2024).

Neurogenesis dynamics in the olfactory bulb:deciphering circuitry organization, function, and adaptive plasticity

Adult neurogenesis persists after birth in the subventricular zone, with new neurons migrating to the granule cell layer and glomerular layers of the olfactory bulb, where they integrate into existing circuitry as inhibitory interneurons. The generation of these new neurons in the olfactory bulb supports both structural and functional plasticity, aiding in circuit remodeling triggered by memory and learning processes. However, the presence of these neurons, coupled with the cellular diversity within the olfactory bulb, presents an ongoing challenge in understanding its network organization and function. Moreover,the continuous integration of new neurons in the olfactory bulb plays a pivotal role in regulating olfactory information processing. This adaptive process responds to changes in epithelial composition and contributes to the formation of olfactory memories by modulating cellular connectivity within the olfactory bulb and interacting intricately with higher-order brain regions. The role of adult neurogenesis in olfactory bulb functions remains a topic of debate. Nevertheless, the functionality of the olfactory bulb is intricately linked to the organization of granule cells around mitral and tufted cells. This organizational pattern significantly impacts output, network behavior, and synaptic plasticity, which are crucial for olfactory perception and memory. Additionally, this organization is further shaped by axon terminals originating from cortical and subcortical regions. Despite the crucial role of olfactory bulb in brain functions and behaviors related to olfaction, these complex and highly interconnected processes have not been comprehensively studied as a whole. Therefore, this manuscript aims to discuss our current understanding and explore how neural plasticity and olfactory neurogenesis contribute to enhancing the adaptability of the olfactory system. These mechanisms are thought to support olfactory learning and memory, potentially through increased complexity and restructuring of neural network structures, as well as the addition of new granule granule cells that aid in olfactory adaptation. Additionally, the manuscript underscores the importance of employing precise methodologies to elucidate the specific roles of adult neurogenesis amidst conflicting data and varying experimental paradigms. Understanding these processes is essential for gaining insights into the complexities of olfactory function and behavior.

The interaction between KIF21A and KANK1 regulates dendritic morphology and synapse plasticity in neurons

Morphological alterations in dendritic spines have been linked to changes in functional communication between neurons that affect learning and memory.Kinesin-4 KIF21A helps organize the microtubule-actin network at the cell cortex by interacting with KANK1;however,whether KIF21A modulates dendritic structure and function in neurons remains unknown.In this study,we found that KIF21A was distributed in a subset of dendritic spines,and that these KIF21A-positive spines were larger and more structurally plastic than KIF21A-negative spines.Furthermore,the interaction between KIF21A and KANK1 was found to be critical for dendritic spine morphogenesis and synaptic plasticity.Knockdown of either KIF21A or KANK1 inhibited dendritic spine morphogenesis and dendritic branching,and these deficits were fully rescued by coexpressing full-length KIF21A or KANK1,but not by proteins with mutations disrupting direct binding between KIF21A and KANK1 or binding between KANK1 and talin1.Knocking down KIF21A in the hippocampus of rats inhibited the amplitudes of long-term potentiation induced by high-frequency stimulation and negatively impacted the animals’cognitive abilities.Taken together,our findings demonstrate the function of KIF21A in modulating spine morphology and provide insight into its role in synaptic function.

Practice patterns among ophthalmic surgeons in treating concomitant oculoplastic conditions in preoperative period:A questionnaire-based study

BACKGROUND Addressing oculoplastic conditions in the preoperative period ensures both the safety and functional success of any ophthalmic procedure.Some oculoplastic conditions,like nasolacrimal duct obstruction,have been extensively studied,whereas others,like eyelid malposition and thyroid eye disease,have received minimal or no research.AIM To investigate the current practice patterns among ophthalmologists while treating concomitant oculoplastic conditions before any subspecialty ophthalmic intervention.METHODS A cross-sectional survey was disseminated among ophthalmologists all over India.The survey included questions related to pre-operative evaluation,anaesthetic and surgical techniques preferred,post-operative care,the use of adjunctive therapies,and patient follow-up patterns.RESULTS A total of 180 ophthalmologists responded to the survey.Most practitioners(89%)felt that the ROPLAS test was sufficient during pre-operative evaluation before any subspecialty surgery was advised.The most common surgical techniques employed were lacrimal drainage procedures(Dacryocystorhinostomy)(63.3%),eyelid malposition repair(36.9%),and ptosis repair(58.7%).Post-operatively,47.7%of respondents emphasized that at least a 4-week gap should be maintained after lacrimal drainage procedures and eyelid surgeries.Sixty-seven percent of ophthalmologists felt that topical anaesthetic procedures should be preferred while performing ocular surgeries in thyroid eye disease patients.CONCLUSION Approximately 50%of ophthalmologists handle prevalent oculoplastic issues themselves,seeking the expertise of an oculoplastic surgeon under particular conditions.Many ophthalmologists still favor using ROPLAS as a preliminary screening method before proceeding with cataract surgery.Eyelid conditions and thyroid eye disease are not as commonly addressed before subspecialty procedures compared to issues like nasolacrimal duct obstruction and periocular infections.

Microplastic and nanoplastic exposure and risk of diabetes mellitus

The issue of plastic pollutants has become a growing concern.Both microplastics(MPs)(particle size<5 mm)and nanoplastics(NPs)(particle size<1μm)can cause DNA damage,cytotoxicity,and oxidative stress in various organisms.The primary known impacts of microplastic/nanoplastic are observed in the liver and respiratory system,leading to hepatotoxicity and chronic obstructive pulmonary disease.Although research on the effects of MPs and NPs on diabetes is still in its early stages,there are potential concerns.This editorial highlights the risk to diabetics from co-exposure to contaminants and MPs/NPs,supported by evidence from animal studies and the various chemical compositions of MPs/NPs.

基于Seam Carving的逐帧优化视频缩放方法

视频缩放是近年来数字图像处理领域的一项热点问题.针对整个视频进行缩放的方法会带来庞大的内存占用与计算量,导致效率低下、实用性较差;而针对视频中的每一帧进行缩放的方法难以维持视频的时空一致性.为此,基于Seam Carving方法,提出一种逐帧优化的视频缩放方法.首先逐帧读入视频,按照梯度求出当前帧的能量图,并使用高速缓存的置换思想调整能量图;然后根据能量图找出缝;最后使用线性插值的方法删除缝,得到目标大小的帧.实验结果表明,该方法不仅能够在所处理的每一帧中保持图像的重要内容,并且可以维持整个视频的时空一致性,保持较好的视觉效果.

改进的Seam Carving瞬时频率估计算法研究

旋转机械升降速阶段微弱振动信号具有潜在性和动态响应的微弱性等特点,因此瞬时频率是限制早期故障诊断发展的关键问题.针对这一问题,该文引入了图像处理领域的Seam Carving(SC)算法,并对SC算法进行改进,结合短时傅里叶变换(Short-Time Fourier Transformation,STFT),提出了STFT-SC瞬时频率估计算法.STFT-SC算法采用STFT对振动信号进行时频分析,然后利用Seam Carving算法中的能量梯度,采用动态规划的思想寻求seam路径,实现了旋转机械振动信号的一阶瞬时频率提取,最后对一阶时频数据进行时域重构,将重构的时域信号和仿真信号进行相关性分析.结果表明,STFT-SC算法对于高噪声、邻近阶比等振动信号都具有非常好的效果.

一种改进的Seam Carving图像缩放算法

针对图像非比例缩放时出现的失真,基于Seam Carving算法进行了研究:重新定义了显著能量,以突出图像的重要内容;优化了基于能量图的查找方法,通过减少像素线穿越图像中重要内容的次数,以保护图像重要内容不被扭曲;结合双线性插值以限制抽取像素线的数目,以避免图像内容过度抽取。实验结果表明,改进算法能够更好的保护图像中的重要内容、提高图像缩放的质量,并提升Seam Carving算法的执行效率。

Seam Carving和显著性分析的图像缩放方法研究

提出一种基于Seam Carving和显著性分析相结合的图像缩放方法。通过不断移除(插入)缝合线来实现对图像尺寸的缩小(放大)。通过研究Seam Carving在图像缩放及目标保留移除中存在的问题,已有的基于像素点计算图像能量函数的方法会对某些图像失效,提出一种基于显著性区域求解图像能量函数的SDRS方法。实验结果表明,Seam Carving方法对自然图像的处理能够达到令人满意的效果。Seam Carving方法和SDRS方法的结合,也是一个创新点。

基于全局图像结构信息的Seam Carving算法

提出了一种基于全局图像结构信息的Seam Carving算法,它根据像素的重要性修改图像尺寸和比例。通过从图像提取特定方向的边缘结构信息,再利用每个像素的梯度信息,从全局和局部两方面定义新的像素能量计算函数,以此来阻止seam通路与特定方向图像边缘的交叉,避免边缘像素的不一致位移,以此保持图像的边缘结构。实验结果证明,算法减少了处理后图像的结构形变,有效改进了Seam Carving算法的处理效果。

基于Seam Carving的图像放大算法研究与改进

针对Seam Carving图像放大算法中,能量值前k小的像素带可能出现共用点的情况,提出了一种改进的图像放大自适应算法,通过判断图像中出现共用点的像素带的情况及计算其共用率的值,经过一定的策略确定待新增的像素带位置,该算法使得图像中视觉关注非重要信息区域像素带的增加能得到较均匀的分布;实现了人工干预选择视觉关注重要区域的方法,避免了自适应算法本身因误判视觉关注重要区域而带来关注对象发生变形扭曲的情况.实验证明,改进的算法能得到更好的放大效果.

基于Se am Carving的海天线检测算法

海天线在远距离舰船目标检测中具有重要参考价值。目前,主要的海天线检测算法有直线拟合法、基于加权信息熵、Hough直线检测等,这些算法不同程度地存在运算量较大、应用场合受限等缺点。文中通过研究Seam Carving原理,将该原理运用到图像中最高能量线的检测,通过计算图像中各点的梯度幅值,采用动态规划技术搜索最高能量线,实现了海天线的检测。实验结果表明,该算法不仅具有较高的检测率,同时具有适用范围广、运算量小、实时性强等特点。

基于扩展马尔科夫特征的Seam-Carving篡改检测

针对能够用于图像篡改的Seam-Carving技术,提出了一种基于扩展的马尔科夫特征的Seam-Carving篡改识别算法。该算法充分考虑了Seam-Carving操作导致的图像频域特征的变化,将传统的利用马尔科夫转移概率矩阵求取的图像特征和基于扩展的马尔科夫转移概率特征进行融合,而后利用支持向量机进行分类训练,从而达到有效识别基于Seam-Carving的图像篡改。实验结果表明,提出的方案性能优于传统的基于马尔科夫转移矩阵的特征选择方法以及现有的一些该类图像篡改检测方法。

基于最小位移可视差的连续Seam Carving算法在图像缩放中的研究

图像缩放技术要求对图像缩放的同时保证重要信息不丢失且物体边缘不发生扭曲。近年来,Seam Carving及其改进算法得到了广泛的关注和研究。由于采用了离散式最小能量线迭代搜索策略,缩放信息无法在迭代过程中传递导致扭曲现象普遍存在。该文针对上述问题提出最小位移可视差(JND)检测算法,能够有效地检测每一次迭代中出现的潜在扭曲信息。能量权重E_(w)能够将JND信息累加传递给后续的迭代过程,从而抑制缩放过程中的边缘扭曲现象。通过JND算法和能量权重,该文首次将离散的Seam Carving模型转变为连续缩放模型。最后,在公共数据集RetargetMe上与最新的图像缩放算法进行多组对比实验,验证了所提方法的有效性和先进性。

基于Seam Carving的图像自适应缩放算法研究

基于Seam Carving的自适应缩放图像质量依赖于线路能量函数的定义和线路的提取。针对Seam Carving算法中可能同时存在多条累积能量值最小的线路,提出一种改进的线路提取方法。当待提取的线路数量小于图像中存在的等值累积能量值线路数量时,通过增加约束条件,由局部前驱点的选择和全局线路位置偏移差共同决定新的提取线路,即在累积能量值和局部较优均相同的情况下位置偏移差较小的线路优先被提取。实验表明,算法提取的线路定位可较好地应用于各种风格属性的图像缩放,特别是内容相对简单的图片能获得更好的裁剪结果,同时视觉关注重点区域出现扭曲变形的问题也得到改善。

The Cultural Implication of Brick Carving Model Calligraphy in Shanxi Chang's Manor

Focusing on the unique style of brick carving model calligraphy in Shanxi Chang's manor,this research analyses the content and feature of the different parts of brick carving,and discuses the presenting cultural deposits.It reflects the deep Confucian culture of Shanxi Chang's manor,with the peculiar charming in culture and aesthetic value.

基于seam carving的内容感知图像缩放算法研究

图像在变换成目标尺寸的缩放过程中，均匀缩放容易造成图像的失真，直接裁剪方式可能又会造成图像重要内容的缺失，seam carving算法是一种基于内容感知的图像自适应缩放算法，能较好地解决以上问题。该文对seam carving缩放算法进行了研究，首先实现了图像水平和垂直方向上的自适应缩放算法，并与传统的缩放算法进行实验对比分析；然后对seam carving算法进行改进，通过人工干预选择用户感兴趣或不感兴趣区域，来更好地满足用户所需的缩放效果。