Search

Abstract Traffic sign detection is critical for autonomous driving technology. However, accurately detecting traffic signs in complex traffic environments remains challenge despite the widespread use of one‐stage detection algorithms known for their real‐time processing capabilities. In this paper, the authors propose a traffic sign detection method based on YOLO v8. Specifically, this study introduces the Space‐to‐Depth (SPD) module to address missed detections caused by multi‐scale variations of traffic signs in traffic scenes. The SPD module compresses spatial information into depth channels, expanding the receptive field and enhancing the detection capabilities for objects of varying sizes. Furthermore, to address missed detections caused by complex backgrounds such as trees, this paper employs the Select Kernel attention mechanism. This mechanism enables the model to dynamically adjust its focus and more effectively concentrate on key features. Additionally, considering the uneven distribution of training data, the authors adopted the WIoUv3 loss function, which optimizes loss calculation through a weighted approach, thereby improving the model's detection performance across various sizes and frequencies of instances. The proposed methods were validated on the CCTSDB and TT100K datasets. Experimental results demonstrate that the authors’ method achieves substantial improvements of 3.2% and 5.1% on the mAP50 metric compared to YOLOv8s, while maintaining high detection speed, significantly enhancing the overall performance of the detection system. The code for this paper is located at https://github.com/dusongjie/TSD‐YOLO‐Small‐Traffic‐Sign‐Detection‐Based‐on‐Improved‐YOLO‐v8

Abstract The limited of texture details information in low‐resolution facial or eye images presents a challenge for gaze estimation. To address this, FSKT‐GE (feature maps similarity knowledge transfer for low‐resolution gaze estimation) is proposed, a gaze estimation framework consisting of both a high resolution (HR) network and low resolution (LR) network with the identical structure. Rather than mere feature imitation, this issue is addressed by assessing the cosine similarity of feature layers, emphasizing the distribution similarity between the HR and LR networks. This enables the LR network to acquire richer knowledge. This framework utilizes a combination loss function, incorporating cosine similarity measurement, soft loss based on probability distribution difference and gaze direction output, along with a hard loss from the LR network output layer. This approach on low‐resolution datasets derived from Gaze360 and RT‐Gene datasets is validated, demonstrating excellent performance in low‐resolution gaze estimation. Evaluations on low‐resolution images obtained through 2×, 4×, and 8× down‐sampling are conducted on two datasets. On the Gaze360 dataset, the lowest mean angular errors of 10.97°, 11.22°, and 13.61° were achieved, while on the RT‐Gene dataset, the lowest mean angular errors of 6.73°, 6.83°, and 7.75° were obtained.

With the development of generative model, the cost of facial manipulation and forgery is becoming lower and lower. Fraudulent data has brought numerous hidden threats in politics, privacy, and cybersecurity. Although many methods of face forgery detection focus on the learning of high frequency forgery traces and achieve promising performance, these methods usually learn features in spatial and frequency independently. In order to combine the information of the two domains, a combined spatial and frequency dual stream network is proposed for face forgery detection. Concretely, a cross self-attention (CSA) module is designed to improve frequency feature interaction and fusion at different scales. Moreover, to augment the semantic and contextual information, frequency guided spatial feature extraction module is proposed to extract and reconstruct the spatial information. These two modules deeply mine the forgery traces via a dual-stream collaborative network. Through comprehensive experiments on different datasets, we demonstrate the effectiveness of proposed method for both within and cross datasets.

Invisible orthodontic treatment is an effective form of malocclusion treatment favored in recent years. The magnitude of its orthodontic force has a crucial impact on the outcome of the treatment and has gained a high level of clinical interest. However, there are very few explorations of in vivo measurements of orthodontic force, and existing studies are limited to a large number of couplings, which are inconvenient for clinical use. In this work, we developed a wireless flexible measurement system that allows quantitative measurement of the orthodontic force of an invisible aligner on a dental model. The system is wireless, tiny, flexible, fast responding, and has a range suitable for the range of orthodontic forces. We show the difference in the orthodontic force applied to different tooth positions and the difference in the orthodontic force applied to different positions of the same tooth. In addition, the system can evaluate the mechanical differences between aligners of different brands and materials as well as the deviation of fabrication results. This system provides a test tool and evaluation method for future real-time assessment of clinical orthodontic forces.

Optical frequency comb with evenly spaced lines over a broad bandwidth has revolutionized the fields of optical metrology and spectroscopy. Here, we propose a fast interleaved dual-comb spectroscopy with sub-femtometer-resolution and absolute frequency, in which two electro-optic frequency combs are swept. Electrically-modulated stabilized laser enables ultrahigh resolution of 0.16 fm (or 20 kHz in optical frequency) and single-shot measurement in 90 ms. Total 20 million points are recorded spanning 3.2 nm (or 400 GHz) bandwidth, corresponding to a spectral sampling rate of 2.5 × 108 points/s under Nyquist-limitation. Besides, considering the trade-off between the measurement time and spectral resolution, a fast single-shot measurement is also realized in 1.6 ms with 8 fm (or 1 MHz) resolution. We demonstrate the 25-averaged result with 30.6 dB spectral measurement signal-to-noise ratio (SNR) by reducing the filter bandwidth in demodulation. The results show great prospect for precise measurement with flexibly fast refresh time, high spectral resolution, and high SNR.

Image dehazing is one of the problems that need to be solved urgently in the field of computer vision. In recent years, more and more algorithms have been applied to image dehazing and achieved good results. However, the image after dehazing still has color distortion, contrast and saturation disorder, and other challenges; in order to solve these problems, in this paper, an effective image dehazing method is proposed, which is based on improved color channel transfer and multiexposure image fusion to achieve image dehazing. First, the image is preprocessed using a color channel transfer method based on k-means. Second, gamma correction is introduced on the basis of guided filtering to obtain a series of multiexposure images, and the obtained multiexposure images are fused into a dehazed image through a Laplacian pyramid fusion scheme based on local similarity of adaptive weights. Finally, contrast and saturation corrections are performed on the dehazed image. Experimental verification is carried out on synthetic dehazed images and natural dehazed images, and it is verified that the method proposed is superior to existing dehazed algorithms from both subjective and objective aspects.

Abstract Background Early diagnosis and continuous monitoring are the key to emergency treatment and intensive care of patients with acute ischemic stroke (AIS). Nevertheless, there has not been a fully accepted method targeting continuous assessment of AIS in clinical. Methods Near-field coupling (NFC) sensing can obtain the conductivity related to the volume of intracranial components with advantages of non-invasiveness, strong penetrability and real-time monitoring. In this work, we built a multi-parameter monitoring system that is able to measure changes of phase and amplitude in the process of electromagnetic wave (EW) reflection and transmission. For investigating its feasibility in AIS detection, 16 rabbits were chosen to establish AIS models by bilateral common carotid artery ligation and then were enrolled for monitoring experiments. Results During the 6 h after AIS, the reflection amplitude (RA) shows a decline trend with a range of 0.69 dB and reflection phase (RP) has an increased variation of 6.48° . Meanwhile, transmission amplitude (TA) and transmission phase (TP) decrease 2.14 dB and 24.29° , respectively. The statistical analysis illustrates that before ligation, 3 h after ligation and 6 h after ligation can be effectively distinguished by the four parameters individually. When all those parameters are regarded as recognition features in back propagation (BP) network, the classification accuracy of the three different periods reaches almost 100%. Conclusion These results prove the feasibility of multi-parameter NFC sensing to assess AIS, which is promised to become an outstanding point-of-care testing method in the future.

Heatmap-based traditional approaches for estimating human pose usually suffer from drawbacks such as high network complexity or suboptimal accuracy. Focusing on the issue of multi-person pose estimation without heatmaps, this paper proposes an end-to-end, lightweight human pose estimation network using a multi-scale coordinate attention mechanism based on the Yolo-Pose network to improve the overall network performance while ensuring the network is lightweight. Specifically, the lightweight network GhostNet was first integrated into the backbone to alleviate the problem of model redundancy and produce a significant number of effective feature maps. Then, by combining the coordinate attention mechanism, the sensitivity of our proposed network to direction and location perception was enhanced. Finally, the BiFPN module was fused to balance the feature information of different scales and further improve the expression ability of convolutional features. Experiments on the COCO 2017 dataset showed that, compared with the baseline method YOLO-Pose, the average accuracy of the proposed network on the COCO 2017 validation dataset was improved by 4.8% while minimizing the amount of network parameters and calculations. The experimental results demonstrated that our proposed method can improve the detection accuracy of human pose estimation while ensuring that the model is lightweight.

We propose a novel generalized linear optical sampling (GLOS) technique realized by using an electro-optic frequency comb (EOFC) as the sampling signal. GLOS technique is demonstrated as a bandwidth compression process in frequency domain instead of gating effect in time domain. An EOFC without the limitation to be ultra-short pulse serves as sampling signal is pre-measured. In experiments, the waveforms are sampled by an EOFC with agile repetition rates and bandwidths. After a demodulation process with pre-measured information, the original signal under test in both intensity and phase fields can be recovered. The results obtained from the proposed method are consistent with those from traditional linear optical sampling technique. Besides, with a high average mode power, EOFC-based GLOS technique realizes more than 10 dB SNR improvement and has ability to detect weak signal with a power of -47.3 dBm. Our demonstration opens the way for cost-effective comb sources to be used in optical sampling fields.

Objective: Laryngopharyngeal Reflux Disease (LPRD) is prevalent and has a range of symptoms. However, diagnosing LPRD is difficult because of the lack of specific symptoms or clinical gold standard. An objective and reliable test, which does not rely on a perfect clinical gold standard, is required in clinical practice. Methods: 60 normal volunteers and 74 confirmed Laryngopharyngeal Reflux (LPR) patients were labelled based on the combined consideration of the reflux symptom index, reflux finding score, 24h oropharyngeal pH monitoring and results of anti-reflux treatment. 72 candidate features were extracted from pH recordings and the most efficient feature combination was detected using a stepwise wrapper method. The labelled data were combined with 1552 unlabelled data for feature selection and model training using semi-supervised learning. A latent class model method was used to assess the proposed model based on 64 additional validation data and an imperfect clinical reference test. Results: A new score (named W score), which significantly improved the sensitivity of the LPRD test (82.67% vs. 24.09%) and has a relatively high specificity (80.19%), was proposed. W score concurs with the complicated clinical test. Conclusion: W score which significantly improves LPRD diagnostic efficiency, can aid the clinical diagnosis of LPRD. W score provides an objective, efficient, and reliable indicator for the application of anti-reflux treatment in clinical practice. Significance: More LPRD patients can benefit from anti-reflux treatment in clinical practice and fewer patients may suffer from, for example, the side effects of unnecessary long-term Proton Pump Inhibitors (PPI) treatments.

Melanocortin receptors (MCRs) and their accessory proteins (MRA Ps) evolutionarily first appear in the genome of sea lamprey. The most ancient melanocortin system consists of only two melanocortin receptors (slMCa and slMCb) and one MRAP2 (slMRAP2) protein, but the physiological roles have not been fully explored in this primitive species. Here, we synthesize and characterize the pharmacological features of slMRAP2 protein on two slMCRs. Our results show that the slMRAP2 protein lacks the long carboxyl terminus; it directly interacts and decreases the surface expression but enhances the α-MSH-induced agonism of slMCa and slMCb. In comparison with higher organisms such as elephant shark and zebrafish, we also demonstrate the constantly evolving regul atory function of the carboxyl terminus of MRAP2 protein, the unique antiparallel topology of slMRAP2 dimer and the homo- and hetero-dimerization of two slMCRs. This study elucidates the presence and modulation of melanocortin receptor by the accessory protein of the agnathans for the first time, which provides a better insight of the melanocor tin system in ancient species of chordates.

To investigate the crosstalk between cartilage and fibroblast-like synoviocytes (FLS) in rheumatoid arthritis (RA), we adopted an in vitro coculture system model of collagen-induced arthritis (CIA) cartilage and CIA FLS monolayer. CIA rat samples of the synovium and femur head were collected for isolation of FLS and coculture system. Cartilages were treated with vehicle (Ctrl group), 10 ng/mL interleukin- (IL-) 1α (IL-1α group), and 10 ng/mL IL-1α plus 10 μM dexamethasone (Dex group) for 3 days before coculture with FLS for further 2 days. After the coculture, FLS were collected to determine the influences of articular cartilage on synoviocytes. Whether the CypB-CD147 signaling pathway is involved in the interactions between cartilage and FLS is assayed. Results showed that IL-1α-stimulated CIA cartilage promoted the proliferation and reduced the apoptosis of FLS. Increased inflammatory cytokines and decreased p57 expression were found in cocultured FLS stimulated by IL-1α-challenged CIA cartilage. Upregulation of NF-κB and I-κB kinase β (IKK-β) and downregulation of the inhibitor of NF-κBα (I-κBα) protein were observed in cocultured FLS. After coculture, significant increases in the expression of cyclophilin B (CypB) and CD147 were observed in CIA cartilage and FLS, respectively. Furthermore, results of immunofluorescence staining showed that the anti-CD147 antibody significantly suppressed p65 nuclear translocation in cocultured FLS stimulated by IL-1α-challenged CIA cartilage. In conclusion, inflammatory effects in the cartilage-FLS coculture system are associated with the CypB-CD147 mediating NF-κB pathway which may further enhance the inflammation in RA.

Hepatoid adenocarcinoma of lung (HAL) is a rare and aggressive tumor. The current study reported a new HAL case in the right lower lung with high serum α-fetoprotein (AFP) level in a 71-year-old male patient. After the confirmation of morphology and immunohistochemistry, the patient was diagnosed clinically with HAL and treated with radio-frequency ablation. However, the patient whose disease progressed eventually died 4 months after diagnosis. Whole genome sequencing analysis identified a driver gene mutation in the FAT atypical cadherin 1 gene (FAT1) and the copy number loss. The tumor was microsatellite-stable and tumor mutation burden (TMB) was 1.69 mutations/Mb. PD-L1 expression was negative by IHC. Our finding provide further clues for the molecular basis of HAL and the efficacy of immunotherapy needs to be explored.

Disregulation of fatty acid oxidation, one of the major mechanisms for maintaining hepatic lipid homeostasis under fasting conditions, leads to hepatic steatosis. Although obesity and type 2 diabetes-induced endoplasmic reticulum (ER) stress contribute to hepatic steatosis, it is largely unknown how ER stress regulates fatty acid oxidation. Here we show that fasting glucagon stimulates the dephosphorylation and nuclear translocation of histone deacetylase 5 (HDAC5), where it interacts with PPARα and promotes transcriptional activity of PPARα. As a result, overexpression of HDAC5 but not PPARα binding-deficient HDAC5 in liver improves lipid homeostasis, whereas RNAi-mediated knockdown of HDAC5 deteriorates hepatic steatosis. ER stress inhibits fatty acid oxidation gene expression via calcium/calmodulin-dependent protein kinase II-mediated phosphorylation of HDAC5. Most important, hepatic overexpression of a phosphorylation-deficient mutant HDAC5 2SA promotes hepatic fatty acid oxidation gene expression and protects against hepatic steatosis in mice fed a high-fat diet. We have identified HDAC5 as a novel mediator of hepatic fatty acid oxidation by fasting and ER stress signals, and strategies to promote HDAC5 dephosphorylation could serve as new tools for the treatment of obesity-associated hepatic steatosis.

In a coherent diffraction imaging (CDI) system, the information of the sample is retrieved from the diffraction patterns recorded by the image sensor via multiple iterations. The limited dynamic range of the image sensor restricts the resolution of the reconstructed sample information. To alleviate this problem, the high dynamic range imaging technology is adopted to increase the signal-to-noise ratio of the diffraction patterns. A sequence of raw diffraction images with differently exposure time are recorded by the image sensor. Then, they are fused to generate a high quality diffraction pattern based on the response function of the image sensor. With the fused diffraction patterns, the resolution of the coherent diffraction imaging can be effectively improved. The experiments on USAF resolution card is carried out to verify the effectiveness of our proposed method, in which the spatial resolution is improved by 1.8 times using the high dynamic range imaging technology.

ENHANCED DISEASE RESISTANCE1 (EDR1) encodes a Raf-like mitogen-activated protein kinase, and it acts as a negative regulator of disease resistance and ethylene-induced senescence. Mutations in the EDR1 gene can enhance resistance to powdery mildew both in monocotyledonous and dicotyledonous plants. However, little is known about EDR1-like gene members from a genome-wide perspective in plants. In this study, the tobacco (Nicotiana tabacum) EDR1-like gene family was first systematically analyzed. We identified 19 EDR1-like genes in tobacco, and compared them to those from Arabidopsis, tomato and rice. Phylogenetic analyses divided the EDR1-like gene family into six clades, among them monocot and dicot plants were respectively divided into two sub-clades. NtEDR1-1A and NtEDR1-1B were classified into clade I in which the other members have been reported to negatively regulate plant resistance to powdery mildew. The expression patterns of tobacco EDR1-like genes were analyzed after plants were challenged by Golovinomyces orontii, and showed that several other EDR1-like genes were induced after infection, as well as NtEDR1-1A and NtEDR1-1B. Expression analysis showed that NtEDR1-13 and NtEDR1-16 had exclusively abundant expression patterns in roots and leaves, respectively, and the remaining NtEDR1-like members were actively expressed in most of the tissue/organ samples investigated. Our findings will contribute to further study of the physiological functions of EDR1-like genes in tobacco.

Strigolactones (SLs) are a class of phytohormones that regulate plant architecture. Carotenoid cleavage dioxygenase (CCD) genes are involved in the biosynthesis of SLs and are identified and characterized in many plants. However, the function of CCD genes in tobacco remains poorly understood. In this study, two closely related genes NtCCD8A and NtCCD8B were cloned from tobacco (Nicotiana tabacum L.). The two NtCCD8 genes are orthologues of the tomato (Solanum lycopersicum) carotenoid cleavage dioxygenase 8 (SlCCD8) gene. NtCCD8A and NtCCD8B were primarily expressed in tobacco roots, but low expression levels of these genes were detected in all plant tissues, and their transcript levels significantly increased in response to phosphate limitation. NtCCD8A and NtCCD8B mutations were introduced into tobacco using the CRISPR/Cas9 system and transgenic tobacco lines for both ntccd8 mutant alleles were identified. The ntccd8a and ntccd8b mutant alleles were inactivated by a deletion of three nucleotides and insertion of one nucleotide, respectively, both of which led to the production of premature stop codons. The ntccd8 mutants had increased shoot branching, reduced plant height, increased number of leaves and nodes, and reduced total plant biomass compared to wild-type plants; however, the root-to-shoot ratio was unchanged. In addition, mutant lines had shorter primary roots and more of lateral roots than wild type. These results suggest that NtCCD8 genes are important for changes in tobacco plant architecture.

OBJECTIVE: Collagen peptides (CP) compounds, as bone health supplements, are known to play a role in the treatment of osteoporosis. However, the molecular mechanisms of this process remain unclear. This study aimed to investigate the effects of bovine CP compounds on the proliferation and differentiation of MC3T3-E1 cells. METHODS: Mouse pre-osteoblast cell line MC3T3-E1 subclone 4 cells were treated with bovine CP compounds. Cell proliferation was analyzed by MTT assays and the cell cycle was evaluated by flow cytometry scanning. Furthermore, MC3T3-E1 cell differentiation was analyzed at the RNA level by real-time PCR and at the protein level by western blot analysis for runt-related transcription factor 2 (Runx2), a colorimetric p-nitrophenyl phosphate assay for alkaline phosphatase (ALP), and ELISA for osteocalcin (OC). Finally, alizarin red staining for mineralization was measured using Image Software Pro Plus 6.0. RESULTS: Cell proliferation was very efficient after treatment with different concentrations of bovine CP compounds, and the best concentration was 3 mg/mL. Bovine CP compounds significantly increased the percentage of MC3T3-E1 cells in G2/S phase. Runx2 expression, ALP activity, and OC production were significantly increased after treatment with bovine CP compounds for 7 or 14 days. Quantitative analyses with alizarin red staining showed significantly increased mineralization of MC3T3-E1 cells after treatment with bovine CP compounds for 14 or 21 days. CONCLUSIONS: Bovine CP compounds increased osteoblast proliferation, and played positive roles in osteoblast differentiation and mineralized bone matrix formation. Taking all the experiments together, our study indicates a molecular mechanism for the potential treatment of osteoarthritis and osteoporosis.