Asymmetric Loss Based on Image Properties for Deep Learning-Based Image Restoration

Deep learning techniques have significantly improved image restoration tasks in recent years.As a crucial compo-nent of deep learning,the loss function plays a key role in network optimization and performance enhancement.However,the currently prevalent loss functions assign equal weight to each pixel point during loss calculation,which hampers the ability to reflect the roles of different pixel points and fails to exploit the image’s characteristics fully.To address this issue,this study proposes an asymmetric loss function based on the image and data characteristics of the image recovery task.This novel loss function can adjust the weight of the reconstruction loss based on the grey value of different pixel points,thereby effectively optimizing the network training by differentially utilizing the grey information from the original image.Specifically,we calculate a weight factor for each pixel point based on its grey value and combine it with the reconstruction loss to create a new loss function.This ensures that pixel points with smaller grey values receive greater attention,improving network recovery.In order to verify the effectiveness of the proposed asymmetric loss function,we conducted experimental tests in the image super-resolution task.The experimental results show that the model with the introduction of asymmetric loss weights improves all the indexes of the processing results without increasing the training time.In the typical super-resolution network SRCNN,by introducing asymmetric weights,it is possible to improve the peak signal-to-noise ratio(PSNR)by up to about 0.5%,the structural similarity index(SSIM)by up to about 0.3%,and reduce the root-mean-square error(RMSE)by up to about 1.7%with essentially no increase in training time.In addition,we also further tested the performance of the proposed method in the denoising task to verify the potential applicability of the method in the image restoration task.

成冰纪南华盆地海水氧化还原性质的钨稳定同位素限定

成冰纪是大气和海洋氧化还原状态的关键转折期.重建成冰纪海洋氧化还原历史对解析早期动物演化、氧化还原敏感元素地球化学循环等诸多问题均具有重要意义.本研究针对具有重要全球对比意义和较好研究基础的南华盆地,尝试利用新近开发的钨稳定同位素体系,限定成冰纪间冰期陆架边缘海氧化还原环境.结果表明,南华盆地成冰纪间冰期黑色页岩层位的δ^(186)/^(184)W值明显高于陆源碎屑,指示具有较高δ^(186)/^(184)W值组分的加入.根据现代氧化海洋自生沉积钨稳定同位素组成,本研究认为,较高δ^(186)/^(184)W值端元来自海洋自生沉积组分.由于硫化海水中钨主要以硫代钨酸根形式存在,而硫代钨酸根具有较高的溶解度,硫化底层海水形成的沉积物中难以保存自生钨沉积组分.因此,钨稳定同位素组成表明,南华盆地成冰纪间冰期底层海水至少存在间歇性非硫化状态.结合已发表的铁组分和其他海水环境地球化学指标,本研究认为,南华盆地与当今波罗的海类似,间歇性通风导致底层硫化海水局部氧化,从而形成自生钨沉积.本研究展示了钨稳定同位素限定古海水氧化还原状态的巨大潜力.

Potentiating CD8^(+) T cell antitumor activity by inhibiting PCSK9 to promote LDLR-mediated TCR recycling and signaling

Metabolic regulation has been proven to play a critical role in T cell antitumor immunity.However,cholesterol metabolism as a key component of this regulation remains largely unexplored.Herein,we found that the low-density lipoprotein receptor(LDLR),which has been previously identified as a transporter for cholesterol,plays a pivotal role in regulating CD8+T cell antitumor activity.Besides the involvement of cholesterol uptake which is mediated by LDLR in T cell priming and clonal expansion,we also found a non-canonical function of LDLR in CD8+T cells:LDLR interacts with the T-cell receptor(TCR)complex and regulates TCR recycling and signaling,thus facilitating the effector function of cytotoxic T-lymphocytes(CTLs).Furthermore,we found that the tumor microenvironment(TME)downregulates CD8+T cell LDLR level and TCR signaling via tumor cell-derived proprotein convertase subtilisin/kexin type 9(PCSK9)which binds to LDLR and prevents the recycling of LDLR and TCR to the plasma membrane thus inhibits the effector function of CTLs.Moreover,genetic deletion or pharmacological inhibition of PCSK9 in tumor cells can enhance the antitumor activity of CD8+T cells by alleviating the suppressive effect on CD8+T cells and consequently inhibit tumor progression.While previously established as a hypercholesterolemia target,this study highlights PCSK9/LDLR as a potential target for cancer immunotherapy as well.

Hexagonal boron nitride/poly(vinyl butyral)composite coatings for corrosion protection of copper

Hexagonal boron nitride(h-BN)fillers are incorporated into poly(vinyl butyral)(PVB)coatings to improve the corrosion protection performance of copper.It has been revealed that the h-BN fillers exhibit an excellent dispersiblility in PVB coating due to the non-covalent interactions between h-BN fillers and the PVB molecules.Electrochemical characterization reveals that the corrosion resistance of the BN-reinforced PVB(BN-P)coating is 5-6 orders of magnitude higher than that of the pristine PVB coating.Photographs and metallography show that the copper substrate beneath the BN-P coating does not suffer from corrosion after immersion for 2 months,indicating that the BN-P coating can provide a long-term protective barrier for the underlying copper substrate.Loading 0-0.25 g h-BN fillers in 2.0 g PVB,the corrosion protection performance increases with increasing the loading of h-BN fillers.The scratch test results suggest that h-BN fillers do not accelerate copper corrosion when the BN-P coating is damaged.

Preparation of Ag@CuFe_(2)O_(4)@TiO_(2) nanocomposite films and its performance of photoelectrochemical cathodic protection

A Ag@CuFe_(2)O_(4)@TiO_(2) nanocomposite film with high performance of photogenerated cathodic protection was prepared by hydrothermal and photoreduction methods.The results showed that when the CuFe_(2)O_(4) hydrothermal reaction time was 6 h and the AgNO_(3) concentration was 0.1 M,the Ag@CuFe_(2)O_(4)@TiO_(2) nanocomposite material performed the best cathodic protection capability for 304 stainless steel(304SS).In this case,the protective potential achieved-930 mV(versus SCE)associated with the photocurrent density of 475μA/cm^(2),which was 14.8 times that of pure TiO_(2) nanowires.In the dark,the nanocomposite provided cathodic protection of up to 485 mV for 304SS.Due to the heterogeneous junctions at the two interfaces among the three kinds of nanocomposite materials,the build-in electric field was fabricated,which promoted the separation efficiency of photogenerated electrons and holes and effectively improved the photochemical cathodic protection of 304SS.

Correction to:Potentiating CD8^(+)T cell antitumor activity by inhibiting PCSK9 to promote LDLR-mediated TCR recycling and signaling

Figure 1.LDLR deficiency hinders the antitumor activity of CD8^(+)T cells.(A)Transcriptional level of genes involved in cholesterol transport in naïve CD8^(+)T cells,CTLs and CD8^(+)TILs(isolated at Day3 or Day7 post CTLs adoptive transfer),(n=4).(B)LDLR expression level on CTLs and CD8^(+)TILs(isolated at Day3 post CTLs adoptive transfer),(n=4).

Three‑Dimensional Nano‑displacement Measurement by Four‑Beam Laser Interferometry

A 3D nano-displacement measurement method,where the difference in phase between the beams in a four-beam laser interference is changed,is proposed.Simulation results demonstrate that the variation of phase difference causes the deviation of the interference pattern in the laser interference system.Based on this theory,we design and build a four-beam laser interference system.The corner cube prism in the optical path is shifted,and the phase of the beam is changed by applying different voltages to a piezoelectric stage.The phase difference is obtained by analyzing the lattice pattern with subpixel precision,and then the displacement is determined by correlation operation.The experimental measurement results are consistent with the theoretical analysis,thereby verifying the feasibility of this measurement method.