Adaptive Feature Research Articles

DMDLK-Net: A dynamic multi-scale feature fusion network with deformable large kernel for medical segmentation

Abstract. Skin disease image segmentation is a crucial component of computer-aided diagnosis, providing precise localization and delineation of lesions that enhance diagnostic accuracy and efficiency. Despite significant advancements in convolutional neural networks (CNNs), there remains substantial room for improvement in segmentation performance due to the diverse and complex nature of skin lesions. In this study, we propose DMDLK-Net, a dynamic multi-scale feature fusion network with deformable large kernels, specifically designed to address the challenges in skin disease segmentation. Our network incorporates a Dynamic Deformable Large Kernel (DDLK) module and a Dynamic Multi-Scale Feature Fusion (DMFF) module, enhancing the model's ability to capture intricate lesion features. We present the performance of DMDLK-Net on the ISIC-2018 dataset, highlighting its promising results. Key contributions of this work include the innovative use of deformable large kernels for adaptive feature extraction and the introduction of dynamic multi-scale fusion to balance local and global information. Our experimental results confirm the effectiveness of DMDLK-Net in delivering high-precision segmentation, thus providing a reliable tool for clinical applications.

Toxigenic effects of sponges and benthic diatoms on marine invertebrates and their possible biotechnological applications

Secondary metabolites play important physiological roles being bioactive as defences against other organisms, or attractive signals used for various purposes, including reproduction. Their production and the emission in the environment may be viewed as an adaptive feature subjected to evolutionary selection. They were demonstrated to be useful for applications in various biotechnological fields, such as pharmaceutical, nutraceutical and cosmeceutical. Sponges and microalgae, including diatoms, are the most promising sources of bioactive compounds from the sea. We aimed at detecting the ecotoxicological effects of crude extracts and fractions obtained from three marine sponges, Geodia cydonium, Haliclona (Halichoclona) vansoesti and Agelas oroides and two benthic diatoms, Nanofrustulum shiloi and Cylindrotheca closterium on model marine organisms. We tested their effects on the Mediterranean purple sea urchin, Paracentrotus lividus, and on two diatoms, Phaeodactylum tricornutum and Cylindrotheca closterium, chosen because they are considered standard indicators for assessment of ecological impacts. Our results showed that extracts and fractions from both sponges and diatoms may be harmful for model invertebrates. However, eggs appeared “protected” from sponge allelochemicals when still unfertilized. The majority of sponge fractions exhibited noticeable impacts during the post-fertilization treatments. In contrast, fractions from diatoms notably increased the rate of malformations compared to the control, both in pre- and post-fertilization treatments.

MM-IRSTD: Conv Self-Attention-Based Multi-Modal Small and Dim Target Detection in Infrared Dual-Band Images

Infrared multi-band small and dim target detection is an important research direction in the fields of modern remote sensing and military surveillance. However, achieving high-precision detection remains challenging due to the small scale, low contrast of small and dim targets, and their susceptibility to complex background interference. This paper innovatively proposes a dual-band infrared small and dim target detection method (MM-IRSTD). In this framework, we integrate a convolutional self-attention mechanism module and a self-distillation mechanism to achieve end-to-end dual-band infrared small and dim target detection. The Conv-Based Self-Attention module consists of a convolutional self-attention mechanism and a multilayer perceptron, effectively extracting and integrating input features, thereby enhancing the performance and expressive capability of the model. Additionally, this module incorporates a dynamic weight mechanism to achieve adaptive feature fusion, significantly reducing computational complexity and enhancing the model’s global perception capability. During model training, we use a spatial and channel similarity self-distillation mechanism to drive model updates, addressing the similarity discrepancy between long-wave and mid-wave image features extracted through deep learning, thus improving the model’s performance and generalization capability. Furthermore, to better learn and detect edge features in images, this paper designs an edge extraction method based on Sobel. Finally, comparative experiments and ablation studies validate the advancement and effectiveness of our proposed method.

Adaptive Trials in Stroke: Current Use and Future Directions.

Inclusion of adaptive design features in a clinical trial provides preplanned flexibility to dynamically modify a trial during its conduct while preserving validity and integrity. Adaptive trials are needed to accelerate the conduct of more efficient, informative, and ethical clinical research in the field of neurology. Stroke is a natural candidate for adoption of these innovative approaches to trial design. This Research Methods in Neurology article is informed by a scoping review that identified 45 completed or ongoing adaptive clinical trials in stroke that were appraised: 15 trials had published results with or without a published protocol and 30 ongoing trials (14 trials had a published protocol, and 16 trials were registered only). Interventions spanned acute (n = 28), rehabilitation (n = 8), prevention (n = 8), and rehabilitation and prevention (n = 1). A subsample of these trials was selected to illustrate the utility of adaptive design features and discuss why each adaptive feature was incorporated in the design to best achieve the aim; whether each individual feature was used and whether it resulted in expected efficiencies; and any learnings during preparation, conduct, or reporting. We then discuss the operational, ethical, and regulatory considerations that warrant careful consideration during adaptive trial planning and reflect on the workforce readiness to deliver adaptive trials in practice. We conclude that adaptive trials can be designed, funded, conducted, and published for a wide range of research questions and offer future directions to support adoption of adaptive trial designs in stroke and neurologic research more broadly.

A multi-scale feature extraction and fusion-based model for retinal vessel segmentation in fundus images.

In response to the challenge of low accuracy in retinal vessel segmentation attributed to the minute nature of the vessels, this paper proposes a retinal vessel segmentation model based on an improved U-Net, which combines multi-scale feature extraction and fusion techniques. An improved dilated residual module was first used to replace the original convolutional layer of U-Net, and this module, coupled with a dual attention mechanism and diverse expansion rates, facilitates the extraction of multi-scale vascular features. Moreover, an adaptive feature fusion module was added at the skip connections of the model to improve vessel connectivity. To further optimize network training, a hybrid loss function is employed to mitigate the class imbalance between vessels and the background. Experimental results on the DRIVE dataset and CHASE_DB1 dataset show that the proposed model has an accuracy of 96.27% and 96.96%, sensitivity of 81.32% and 82.59%, and AUC of 98.34% and 98.70%, respectively, demonstrating superior segmentation performance.

Ecological and Anatomical Characteristics of riverbank Plants of Water Currents and Waterbodies of the Lower Amur Region

The results of an ecological and anatomical study of plants in the coastal shallows of rivers and lakes o the Lower Amur region are presented in order to identify their resistance and adaptability to the conditions of existence. 7 stenotopic species were studied. It was revealed that the vegetative organs of plants are characterized by a combination of typical hydromorphic and specific adaptive features. Adaptation of species to sandy-silty shallow habitats is carried out due to histological transformations of the main tissues. The narrow specialization of species does not affect the typical structure of plant organs and does not lead to a simplification of their internal structure. The studied species are more characterized by signs of terrestrial micromorphology (sclerification, suberinization and cutinization of tissues) rather than hydrophytic. It is assumed that a specific complex of shallow flora was formed mainly from terrestrial species.

Multi-scale Adaptive Feature Fusion Hashing for Image Retrieval

AbstractThe hash algorithm has the characteristics of high computational speed and low memory consumption, making it well-suited for massive image search tasks. Currently, most methods rely on deep learning for end-to-end feature extraction and hash encoding. These methods use the last layer feature of the model as the semantic feature of the hash encoding image. However, mainstream research has not realized that the features of different stages of the network contain rich image semantic information, which all affect the performance of retrieval. Based on this, we propose a multi-scale adaptive feature fusion hash image retrieval method, which mines more detailed information about the image by introducing adaptive feature fusion modules at different stages of the network, and incorporates shallow features in the final extracted features to help the model understand the image content. In addition, to maintain the similarity of the generated image hash codes and reduce errors in the process of transitioning from original features to hash codes, similarity loss and quantization loss are applied, while fully utilizing the supervised information of dataset labels to get high-quality hash codes. Experimental proof conducted on the CIFAR-10 and NUS-WIDE datasets that this method outperforms other deep learning-based hash image retrieval methods.

Adaptive feature fusion and disturbance correction for accurate remaining useful life prediction of rolling bearings

As a key component in the transmission system of high-speed trains, bearings need to withstand certain loads while rotating at high speeds. Once a failure occurs, it can directly affect the safety of train operations. Therefore, it is of great significance to establish a reliable remaining useful life model to ensure train operation safety. Addressing the problems of redundant features in the existing multi-feature fusion process, which affects diagnostic performance, and the spurious fluctuations in the fusion features, which cause inaccurate determination of the start fault time and lead to low prediction accuracy of remaining useful life, we propose a method based on adaptive feature fusion and the autoregressive integrated moving average model for rolling bearing prediction. Firstly, features from the time domain, frequency domain, and entropy are extracted. A feature selection mechanism with minimum redundancy is constructed to screen the optimal sensitive feature set. Secondly, based on adaptive feature fusion, the optimal sensitive feature set is dynamically fused, and the spurious fluctuations of the health index are corrected using linear regression and the 3σ principle. Next, a bottom-up time series segmentation method is employed to divide the health status of the Improved Health Indicators. Finally, a remaining useful life prediction model based on the autoregressive integrated moving average model is established. This study demonstrates that the proposed method effectively identifies features that are most sensitive to degradation trends, accurately determines the initial fault moment of bearings, and achieves effective prediction of the remaining useful life of bearings.

Research on fault diagnosis of mechanical component in high-speed train based on one-dimensional convolutional block residual channel attention

The yaw damper and the rolling bearing are key components to ensure the safe and reliable operation of high-speed trains. However, fault identification in these components remains challenging, with traditional feature extraction methods often suffering from low diagnostic accuracy. To address these issues, a one-dimensional convolutional block residual channel attention (1DCBRCA) method for fault diagnosis of high-speed train components is introduced. Firstly, the convolutional layer is constructed for feature extraction, and the convolution block attention module is utilized for adaptive feature optimization in both channel and space dimensions. Subsequently, a residual neural network model is established, with channel attention added after each residual block to focus on critical feature information, which is used for adjusting the weight parameters. Furthermore, the proposed method was validated through damper fault experiments and bearing fault experiments. Comparative analysis with state-of-the-art methods demonstrated that the proposed 1DCBRCA approach achieves higher diagnostic accuracy, confirming its potential to enhance fault diagnosis in high-speed train components.

Intelligent prediction of rail corrugation evolution trend based on self-attention bidirectional TCN and GRU

Analyzing the evolution trend of rail corrugation using signal processing and deep learning is critical for railway safety, as current traditional methods struggle to capture the complex evolution of corrugation. This present study addresses the challenge of accurately capturing this trend, which relies significantly on expert judgment, by proposing an intelligent prediction method based on self-attention (SA), a bidirectional temporal convolutional network (TCN), and a bidirectional gated recurrent unit (GRU). First, multidomain feature extraction and adaptive feature screening were used to obtain the optimal feature set. These features were then combined with principal component analysis (PCA) and the Mahalanobis distance (MD) method to construct a comprehensive health indicator (CHI) that reflects the evolution of rail corrugation. A bidirectional fusion model architecture was employed to capture the temporal correlations between forward and backward information during corrugation evolution, with SA embedded in the model to enhance the focus on key information. The outcome was a rail corrugation trend prediction network that combined a bidirectional TCN, bidirectional GRU, and SA. Subsequently, a multi-strategy improved crested porcupine optimizer (CPO) algorithm was constructed to automatically obtain the optimal network hyperparameters. The proposed method was validated with on-site rail corrugation data, demonstrating superior predictive performance compared to other advanced methods. In summary, the proposed method can accurately predict the evolution trend of rail corrugation, offering a valuable tool for on-site railway maintenance.

Enhancing gait recognition in lower limb exoskeletons: adaptive feature selection and random forest with Bayesian optimization

Abstract To improve human-machine cooperation and enhance the accuracy of gait recognition in wearable lower limb exoskeletons, an enhancement method of gait recognition based on adaptive feature selection and an optimized machine learning algorithm was proposed. In this study, surface electromyography (sEMG) signals of rectus femoris, medialis femoris, lateralis femoris, semitendinosus and biceps femoris were recorded during level-ground walking. Then, time domain, frequency domain, time-frequency domain and nonlinear features were extracted. The integrated value of electromyography, variance, root mean square and wavelength were selected as the time domain features, and the mean power frequency and median frequency were selected as the frequency domain features. Wavelet packet energy was selected as the time-frequency domain feature. Nonlinear features including approximate entropy, sample entropy and fuzzy entropy of sEMG were extracted. To reduce feature dimension and improve the calculation efficiency, adaptive feature selection was performed by particle swarm optimization combined with sigmoid function. Then, the feature matrix was determined as the input for a random forest classifier to recognize different gait phases. An adaptive optimization mechanism based on Bayesian optimization was applied to random forest. Compared with random forest, the overall performance of the optimized model was improved. Its accuracy was increased by 3.57%. The feature selection and adaptive optimization mechanisms in gait recognition not only enhance the accuracy of random forest algorithms applied to sEMG for gait prediction, but also facilitate the flexibility and consistency required for lower limb exoskeleton gait control.

THE EFFECT OF THE REHABILITATION AND PHYSICAL THERAPY PROGRAM ON THE FUNCTIONAL STATUS OF POST-STROKE PATIENTS

The article examines the impact of a physical rehabilitation program on the functional state of the musculoskeletal system in elderly patients who have suffered an intracerebral stroke and are in the conditions of a neurorehabilitation center in the early post-stroke period. Before the implementation of the program, post-stroke patients have pronounced walking disorders and a pronounced dependence on external assistance. The aim of the articale is to reveal the effectiveness of the physical rehabilitation program for post-stroke patients in the early recovery period. Results. Features of hemodynamic adaptation in post-stroke patients in the conditions of the rehabilitation center were revealed. Conclusion. As a result of the research, it was found that post-stroke patients had concomitant diseases: chronic heart failure and arterial hypertension in all studied patients, diabetes – in 20% of patients. All these conditions were taken into account during the selection of optimal means of the program physical rehabilitation. After completing the program, patients became more independent and needed less daily assistance, and also mastered such a vital and necessary skill as walking. This is confirmed by tests of functional status (Bartel index, Rivermead) and assessment of independent walking (Hauser walking index). The level of disability according to the Rankin scale decreased by 20%, the Rivermead mobility index increased by 50.6%, the Barthel index of daily activity increased by 13.5%, the Hauser walking index increased by 31.8%, the range of motion in the joints of the limbs also increased, the level of spasticity according to the modified Ashworth scale decreased by 22.5% in the in the upper and lower extremities, the strength according to the MRCS scale increased, in the extremities by 20%. Thus, after completing the rehabilitation program, all patients observed a increase in paretic muscle strength, decreased muscle spasticity tone, and improved range of motion in the joints. Implementation of the physical rehabilitation program resulted in a significant increase in MRCS muscle strength in the limbs by 20% (p < 0.01). After completing the program, patients became more independent and needed less daily assistance, as well as mastered such a vital and necessary skill as walking. This is confirmed by tests of functional status (Bartel index, Rivermead) and assessment of independent walking (Hauser walking index). The level of disability according to the Rankin scale decreased by 20%, the Rivermead mobility index increased by 50.6%, the Barthel daily activity index increased by 13.5%, the Hauser walking index increased by 31.8%, the range of motion in the joints of the upper and of the lower limbs also increased, the level of spasticity according to the modified Ashworth scale decreased by 22.5% in the upper limb and by 20% in the upper and of the lower limbs, the strength according to the MRCS scale increased, in the limbs by 20%. Patients were able to move with little resistance. The effect of the rehabilitation program and physical therapy is evident in the increase in paretic muscle strength, reduction in spasticity, and increase in joint range of motion. Most patients regained the ability to walk, which contributed to increasing their level of independence. In general, the research results obtained after the implementation of the selected means of physical rehabilitation indicate the effectiveness of such a program.

Linguopoetic Analysis of the Perception of Uncle Tom’s Cabin by Harriet Beecher Stowe Novel in Kyrgyz and Russian Cultures

This article analyzes the linguopoetic perception of Uncle Tom’s Cabin by Harriet Beecher Stowe novel in Kyrgyz and Russian cultures. The goal of the study is not only to analyze the linguistic and poetic means used in the translation and adaptation of the novel into Kyrgyz and Russian languages but also to identify the differences and similarities in the interpretation of the text by representatives of both cultures. The paper examines the cultural contexts that influence the perception of the novel and analyzes how these contexts are reflected in the lexical and stylistic features of the translation. Special attention is given to how Kyrgyz and Russian ethnocultural values manifest in the interpretation of the plot and imagery of the work. The study reveals the features of the linguocultural adaptation of the novel and allows conclusions to be drawn about the significance of the linguopoetic approach in comparative literary studies. In the concluding part of this work, we will review the interpretation of Harriet Beecher Stowe's work in contemporary African American studies, focusing primarily on the interpretation of the novel proposed by Henry Louis Gates Jr., a leading historian, and theorist of African American literature, which reflects the radical Afrocentrism of the 1960s and 1990s.

On-demand imidazolidinyl urea-based tissue-like, self-healable, and antibacterial hydrogels for infectious wound care

Bacterial wound infections are a growing challenge in healthcare, posing severe risks like systemic infection, organ failure, and sepsis, with projections predicting over 10 million deaths annually by 2050. Antibacterial hydrogels, with adaptable extracellular matrix-like features, are emerging as promising solutions for treating infectious wounds. However, the antibacterial properties of most of these hydrogels are largely attributed to extrinsic agents, and their mechanisms of action remain poorly understood. Herein we introduce for the first time, modified imidazolidinyl urea (IU) as the polymeric backbone for developing tissue-like antibacterial hydrogels. As-designed hydrogels behave tissue-like mechanical features, outstanding antifreeze behavior, and rapid self-healing capabilities. Molecular dynamics (MD) simulation and density functional theory (DFT) calculation were employed to well-understand the extent of H-bonding and metal-ligand coordination to finetune hydrogels’ properties. In vitro studies suggest good biocompatibility of hydrogels against mouse fibroblasts & human skin, lung, and red blood cells, with potential wound healing capacity. Additionally, the hydrogels exhibit good 3D printability and remarkable antibacterial activity, attributed to concentration dependent ROS generation, oxidative stress induction, and subsequent disruption of bacterial membrane. On top of that, in vitro biofilm studies confirmed that developed hydrogels are effective in preventing biofilm formation. Therefore, these tissue-mimetic hydrogels present a promising and effective platform for accelerating wound healing while simultaneously controlling bacterial infections, offering hope for the future of wound care.

Automatic kidney stone identification: an adaptive feature-weighted LSTM model based on urine and blood routine analysis.

Kidney stones are the most common urinary system diseases, and early identification is of great significance. The purpose of this study was to use routine urine and blood detection indices to build a deep learning (DL) model to identify the presence of kidney stones in the early stage. A retrospective analysis was conducted on patients with kidney stones who were treated at West China Hospital of Sichuan University from January 2020 to June 2023. A total of 1130 individuals presenting with kidney stones and 1230 healthy subjects were enrolled. The first blood and urine laboratory data of participants at our hospital were collected, and the data were divided into a training dataset (80%) and a verification dataset (20%). Additionally, a long short-term memory (LSTM)-based adaptive feature weighting model was trained for the early identification of kidney stones, and the results were compared with those of other models. The performance of the model was evaluated by the area under the subject working characteristic curve (AUC). The important predictive factors are determined by ranking the characteristic importance of the predictive factors. A total of 17 variables were screened; among the top 4 characteristics according to the weight coefficient in this model, urine WBC, urine occult blood, qualitative urinary protein, and microcyte percentage had high predictive value for kidney stones in patients. The accuracy of the kidney stone (KS-LSTM) learning model was 89.5%, and the AUC was 0.95. Compared with other models, it has better performance. The results show that the KS-LSTM model based on routine urine and blood tests can accurately identify the presence of kidney stones. And provide valuable assistance for clinicians to identify kidney stones in the early stage.

Lotus growth and development in artificial water bodies in the conditions of the Stavropol Upland

Relevance. The creation of a botanical collection of the lotus family made it possible to study the biology and adaptive features of a rare relict species of Nelumbo nucifera and its variety in the conditions of the Stavropol Upland.Methods. The main phases of development (the beginning of growth, budding, flowering) were recorded. Statistical processing was carried out by the method of variance analysis according to G.N. Zaitsev.Results. In 2010, Nelumbo nucifera seeds were purchased from the Volga River Delta and planted in a 100-liter container, which was located in a closed ground. From 2010 to 2014, the plant vegetated every year from the second decade of January to November, in November, vegetation was artificially stopped. With such agricultural techniques, the plant had no flowering and fruiting phases. In June 2014, rhizomes were planted and lotus seeds were sown in a pond made of butyl rubber film. In 2017, the first flowering was observed. During the study period, the plants did not reach natural indicators, on average, the diameter of the leaves is less by 18.5 ± 1.6 cm, the flowers are 2 ± 0.1 cm, the height is 1.0–1.5 m. The duration of flowering is 57 ± 4 days, which corresponds to the natural conditions of lotus growth from mid-July to early September. During the years of research, 495 seeds were obtained. Under the influence of abiotic factors, the nut-bearing lotus lives up to 8 years in the botanical garden. Nelumbo nucifera Betsy grows intensively, flowering occurred in the third year after sowing, leaf diameter 57 ± 3 cm, height above the water surface 78 ± 7 cm, flowering duration 73 ± 7 days, which is longer than that of the nut-bearing lotus, full-fledged seeds are formed.

High-Resolution Reconstruction of Temperature Fields Based on Improved ResNet18.

High-precision measurement of temperature value distributions in production scenarios is of great significance for industrial production, but traditional temperature field reconstruction algorithms rely on the design of manual feature extraction methods with high computational complexity and poor generalization ability. In this paper, we propose a high-precision temperature field reconstruction algorithm based on deep learning, using an efficient adaptive feature extraction method for temperature field reconstruction. We design an improved temperature field reconstruction algorithm based on the ResNet18 neural network; introduce the CBAM attention mechanism in the model; and design a feature pyramid, using M-FPN, a multi-scale feature aggregation network fusing PAN and FPN, to make the extracted feature information propagate multi-dimensionally among different layers to improve the feature characterization ability. Finally, the mean square error is used to guide the model to optimize the training so that the model pays more attention to the data and reduces the large error to ensure that the gap between the predicted value and the real value is small. The experimental results show that the reconstruction accuracy of the improved algorithm presented in this paper is significantly better than that of the original algorithm in the case of typical peaked temperature field distributions.

Unlocking the power of immersive learning: The FAIRI instructional design proposition for adaptive immersive virtual reality

Immersive Virtual Reality (IVR) is a promising tool for providing immersive and adaptive learning experiences by simulating realistic work environments and gathering data on learner's interactions to offer personalized instructional content. However, IVR-supported instruction presents unique challenges due to its heightened immersion, including managing the learner's cognitive load and self-regulation. Incorporating adaptivity into IVR can mitigate these challenges by regulating the complexity of the virtual environment, reducing cognitive strain, and maintaining learner focus through real-time feedback for both learners and instructors. However, recent reviews highlight the underutilization of adaptivity within IVR research, making it difficult to inform the design of adaptive IVR instruction.This paper proposes FAIRI, an instructional design for adaptive IVR in domain-specific training. FAIRI synthesizes recent IVR research relevant to the design of adaptivity for IVR through a purposive literature review, guided by the instructional design elements of the Four-Component Instructional Design-model (4C/ID) and Intelligent Tutoring Systems (ITS). The instructional design elements of the 4C/ID-model and ITS are evaluated for their potential viability and further refined for adaptive IVR-supported instruction based on recent IVR studies. Two key design principles are identified for adaptive IVR: limiting IVR to tasks that benefit from its heightened realism and leveraging its adaptive features to manage the intensity of the experience. Additionally, the application of FAIRI's design principles and elements are demonstrated through a forklift operator training case in IVR. FAIRI aims to catalyse further research into adaptivity for IVR by proposing and aligning promising instructional design elements for adaptive IVR instruction.

Flame and smoke detection based on channel shuffling and adaptive spatial feature fusion

Flame and smoke detection based on channel shuffling and adaptive spatial feature fusion

An Improved YOLOv8n Algorithm for Steel Surface Defect Detection

Due to the manufacturing process and other aspects, steel surface defects vary in shape and size, which is a difficult point for defect detection. For this reason, this paper proposes a steel surface defect detection algorithm based on YOLOv8n. The algorithm improves the C2f module of Backbone in this network by adding Context Guided Block (CG block), which can effectively extract the local features, surrounding context and global context, and integrate this information to improve the precision and accuracy of detection. Meanwhile, we also change the detection header of YOLOv8 to Detect_ASFF by introducing the Adaptive Spatial Feature Fusion (ASFF) technique. This improvement effectively filters out the conflicting information, which can allow the model to have a better adaptability at different scales. In this paper, Shape-IoU is adopted as the bounding box loss function to make up for the traditional IoU loss that ignores the intrinsic properties of the bounding box in the calculation process. After optimisation, mAP@0.5 are improved to 67.9% and mAP@0.5~0.95 are improved to 35.1%, which is 2.8% and 1.1% respectively. The improved algorithm in this paper has improved in detection accuracy, which proves the practicality and effectiveness of the algorithm in this paper.