Head Structures Research Articles

Detection of variety and wax bloom of Shaanxi plum during post-harvest handling

The detection of plum variety and wax bloom has extensive applications in the fields of fruit classification and fruit quality assessment. By automating the detection and identification of plum varieties and wax bloom, it is possible to enhance the efficiency and accuracy of variety identification and quality assessment, and reduce manual intervention and misjudgment, thereby improving the market competitiveness of fruits. Currently, many works focus on improving the detection performance of single attribute detection of plum varieties or wax bloom, and it is often necessary to use two models to detect the same plum variety and quality information separately, which leads to inefficient and resource-consuming problems in practical applications. To solve this problem and improve the efficiency of detection, a Multi-Label detection model based on YOLOv7 is proposed. Firstly, the double detection head structure is introduced to improve the prediction ability for two types of attribute features. Then, the loss function suitable for multi-attribute labels is improved, and two classification loss functions are used to optimize the prediction results of the two types of attribute labels, respectively. Finally, a multi-label non-maximum suppression algorithm is proposed to solve the problem of filtering redundant bounding boxes of multi-attribute labels. Experimental results on the plum image dataset show that the proposed Multi-Label YOLOv7 model achieves a mAP@0.5 of 96.2 %, a precision of 94.6 %, and a recall of 89.5 %. The experimental results show that the Multi-Label YOLOv7 model can effectively detect variety and wax bloom attributes and improve the efficiency of multi-attribute label detection. The code and dataset for this experiment can be found at https://github.com/hejinrong/Muti-Label-YOLOv7.

A stable diffusion enhanced YOLOV5 model for metal stamped part defect detection based on improved network structure

Machine-vision-based defect detection for large metal stamping is a fundamental requirement for improving product quality and inspection speed. However, the performance of machine vision in detecting defects is limited by the large variety of stamped products, significant dimensional differences, and long data-acquisition cycles. To overcome these problems, metal stamping detection has been achieved using the improved YOLOV5 model, which features a slim neck and a multiheaded self-attentive mechanism for detecting wrinkles, holes, and cracks in metal stamping. An image-processing module was used to reduce the impact of metal reflections and improve image quality. Stable diffusion improvement was used to augment the dataset to overcome the small dataset size problem and enhance its generalisation capability. In addition, a BotNet structure with a self-attentive mechanism was introduced into the YOLOV5 model backbone to improve the image feature extraction capability. We then optimised the prediction head structure of YOLOV5 to improve the detection speed and accuracy. Ablation experiments were performed to analyse and verify the effectiveness of each module. The results of the ablation experiments show that the mAP of our proposed stable diffusion improvement data enhancement method, and the YOLO-Bot-VOV algorithm, for metal stamped part defect detection reached 98.2 %, and the parameters were reduced by 0.432 million.

Are Macula or Optic Nerve Head Structures Better at Diagnosing Glaucoma? An Answer Using Artificial Intelligence and Wide-Field Optical Coherence Tomography.

We wanted to develop a deep-learning algorithm to automatically segment optic nerve head (ONH) and macula structures in three-dimensional (3D) wide-field optical coherence tomography (OCT) scans and to assess whether 3D ONH or macula structures (or a combination of both) provide the best diagnostic power for glaucoma. A cross-sectional comparative study was performed using 319 OCT scans of glaucoma eyes and 298 scans of nonglaucoma eyes. Scans were compensated to improve deep-tissue visibility. We developed a deep-learning algorithm to automatically label major tissue structures, trained with 270 manually annotated B-scans. The performance was assessed using the Dice coefficient (DC). A glaucoma classification algorithm (3D-CNN) was then designed using 500 OCT volumes and corresponding automatically segmented labels. This algorithm was trained and tested on three datasets: cropped scans of macular tissues, those of ONH tissues, and wide-field scans. The classification performance for each dataset was reported using the area under the curve (AUC). Our segmentation algorithm achieved a DC of 0.94 ± 0.003. The classification algorithm was best able to diagnose glaucoma using wide-field scans, followed by ONH scans, and finally macula scans, with AUCs of 0.99 ± 0.01, 0.93 ± 0.06 and 0.91 ± 0.11, respectively. This study showed that wide-field OCT may allow for significantly improved glaucoma diagnosis over typical OCTs of the ONH or macula. This could lead to mainstream clinical adoption of 3D wide-field OCT scan technology.

YOLO_Bolt: a lightweight network model for bolt detection

Accurate, fast, and intelligent workpiece identification is of great significance to industrial production. To cope with the limited hardware resources of factory equipment, we have made lightweight improvements based on You Only Look Once v5 (YOLOv5) and proposed a lightweight YOLO named YOLO_Bolt. First, ghost bottleneck lightweight deep convolution is added to the backbone module and neck module of the YOLOv5 detection algorithm to reduce the model volume. Second, the asymptotic feature pyramid network is added to enhance the feature utilization ability, suppress interference information, and improve detection accuracy. Finally, the relationship between the loss function and the decoupling head structure was focused on, and the number of decoupling head layers was redesigned according to different tasks to further improve the detection accuracy of the workpiece detection model. We conducted experimental verification on the MSCOCO 2017 dataset and the homemade bolt dataset. The experimental results show that compared with YOLOv5s, the number of model parameters is only 6.8 M, which is half that of the original model. On the MSCOCO 2017 dataset, the mAP increased by 2.4%. FPS increased by 104 frames/s. On the homemade dataset, the mAP 0.5 increased by 4.2%, and our proposed method is 1.2% higher than the latest YOLOv8s. The improved network can provide effective auxiliary technical support for workpiece detection.

Morphological and vascular evidence of glaucomatous damage in myopic guinea pigs with scleral crosslinking

Guinea pigs are often used as models for myopia studies. However, the imaging structure and vasculature of the optic nerve head (ONH) in guinea pigs are tentative. This study investigated morphological parameters and vascular characteristics of the ONH in guinea pigs with form deprivation (FD) myopia before and after scleral crosslinking (CXL), using optical coherence tomography (OCT) and OCT angiography (OCTA). Refractive error, axial length (AL), intraocular pressure (IOP), and OCT-based structural parameters of the ONH were measured at baseline and 3 weeks after the FD + CXL procedure in guinea pigs. The 88 guinea pigs analysed in this study were aged 3 (n = 29), 4 (n = 51), and 5 (n = 8) weeks. The IOP, AL, average and vertical cup-to-disc ratio (C/D), circumpapillary retinal nerve fibre layer, disc area, and cup volume increased at 3 weeks compared to baseline values (all p < 0.001). The refractive error and rim area decreased at 3 weeks compared to baseline values (all p < 0.001). After adjustment for age, IOP was correlated positively with average C/D (p = 0.039) and negatively with rim area (p = 0.009). The severity of blood signal defects was positively associated with the average C/D at 3 weeks (p = 0.027). These findings may facilitate further research on myopia using guinea pigs.

Türkiye's indigenous genetic resource: Muradiye Kelebek pigeon

The current study was carried out to determine the morphological characteristics of the Muradiye kelebek pigeon, which has been reared for many years by breeders in the Muradiye region. The ages of pigeons were classified into two groups: 12-24 months (age group I) and 25-36 months (age group II). The male pigeons had a significantly higher trunk length (P&amp;lt;0.001), head width (P&amp;lt;0.05), beak length (P&amp;lt;0.001), beak depth (P&amp;lt;0.05), thoracic perimeter (P&amp;lt;0.001), and tarsus diameter (P&amp;lt;0.001) compared to the female pigeons. The body weight (P&amp;lt;0.001), chest width (P&amp;lt;0.05), and thoracic perimeter (P&amp;lt;0.01) values of age group II were higher than those of age group I. It was determined that most of the pigeons were grayish blue-eyed (69.90%) and non-crested (76.72%). It was determined that the head structure of the Muradiye kelebek pigeon was similar to the Edremit kelebek, Muradiye donek, Bursa oynarı, Thrace roller, and Alabadem pigeon genotypes. The spotted plumage color of the Muradiye butterfly pigeon and the other three genotypes (jackal plumage in the Thrace roller, speckled plumage in the Edremit kelebek pigeon, and red/black galaca in the Muradiye donek pigeon) may be a common trait. Since these genotypes share some phenotypic characteristics, the phylogenetic relationships between the Muradiye butterfly pigeon and the other five pigeon genotypes (Edremit kelebek, Muradiye donek, Alabadem, Bursa oynarı, and Thrace roller) should be determined by molecular studies.

Method of comprehensive physiotherapy of dental patients after COVID-19

To describe the mechanism of action and clinical effectiveness of the developed combined physiotherapy method, including the induced technique of piracetam iontophoresis on the frontooccipital technic and acupuncture laser therapy in dental patients with complaints of edentulism progression after COVID-19 on the basis of the analysis of single studies on the post-COVID loss of teeth treatment. A number of patients equal 120 who complained of tooth loss after COVID-19 during the past 6 months were examined. The following initial and end points were considered: dental bleeding and inflammation scores, vascular and endothelial dysfunction markers - levels of intercellular adhesion molecules and their receptors (SlCAM-1, SVCAM-1, VEGF-A, ET-1) before and after treatment. Negative correlation between VEGF-A (pg/ml) concentration in peripheral blood serum and sVCAM-1 (ng/ml) level in the examined patients (r=0.4830, p<0.05) and strong inverse correlation between slCAM-1 (ng/ml) level and sVCAM-1 (r=0.7696, p<0.01) have been established. More significant effects after application of the combined induced method on the head's structures and laser acupuncture have been noted than after acupuncture laser exposure and after inducing technique separately, namely in the form of dental inflammation score correction by 1.76 times (p<0.001), decrease of bleeding score by 2.6 (p<0.05), decrease of concentration of SVCAM-1 by 1.7 times and SlCAM-1 by 2 times (p<0.001), increase of endothelin level by 1.7 times as well as the initial low VEGF-A (pg/ml) by 1.5 times (p<0.01). The developed physiotherapeutic complex, which includes laser acupuncture physiotherapy and induced technique of 5% piracetam iontophoresis, can potentially be considered as a physioprophylactic and therapeutic model of post-COVID edentulism.

Novel design and implementation of irregular fractal arrow head structure microstrip antenna for sub 6 GHz 5G applications

Novel design and implementation of irregular fractal arrow head structure microstrip antenna for sub 6 GHz 5G applications

Unsupervised model for structure segmentation applied to brain computed tomography.

This article presents an unsupervised method for segmenting brain computed tomography scans. The proposed methodology involves image feature extraction and application of similarity and continuity constraints to generate segmentation maps of the anatomical head structures. Specifically designed for real-world datasets, this approach applies a spatial continuity scoring function tailored to the desired number of structures. The primary objective is to assist medical experts in diagnosis by identifying regions with specific abnormalities. Results indicate a simplified and accessible solution, reducing computational effort, training time, and financial costs. Moreover, the method presents potential for expediting the interpretation of abnormal scans, thereby impacting clinical practice. This proposed approach might serve as a practical tool for segmenting brain computed tomography scans, and make a significant contribution to the analysis of medical images in both research and clinical settings.

Novel Design and Implementation of Irregular Fractal Arrow Head Structure Microstrip Antenna for sub SUB 6GHz 5G Applications

Novel Design and Implementation of Irregular Fractal Arrow Head Structure Microstrip Antenna for sub SUB 6GHz 5G Applications

Revenue Relation with VAT in Nepal Government Budget Structure

Taxes are major fiscal policy instruments and important government policy tools have an important role in increasing the rate of capital formation and thereby a high rate of economic growth can be achieved. Globalization, privatization and liberalization have brought the huge change in the country budget and economic environment. The budget management of government is one of the important tasks which play very important role in the nation and international environment which influence the technology and changing business environment has posed many challenges before management. This study aims to find out the Nepal government budget status in economic sectors. Descriptive research has been conducted using annual report of economic survey of ministry of finance. Descriptive analysis suggested that is strong relationship between revenue and expenditure management of Nepal government budget. The main objective of this study is to determine the relationship between revenue and expenditure in the study made over the period 2073/74 to 2077/78 taken as a sample year in this research study among the various fiscal years of Nepal government budget. The statistical methods of trend analysis and descriptive statistics like mean, standard deviation, coefficient of variation and hypothesis test were used to analyze the study outcomes.&#x0D; The results of the study indicated the revenue, expenditure and deficits conditions of Nepal government budget. This research articles report sources of revenue and structure of budget heads of Nepalese economic sector. It also reflects the whole revenue and expenditure positions of Nepal government and effective implementation of revenue estimation to manage deficits conditions. The revenue, expenditure and deficits conditions policy and strategies would also helps to manage deficits conditions by research studying.

Optimization of thin-film formation of naphthalene tetracarboxylic diimide derivatives with head and tail structure and application to transistors

Herein, we characterized vacuum-evaporated thin-films of several new naphthalene tetracarboxylic diimide derivatives (C12-NTCDI-R) as n-type organic thin-film transistor (TFT) materials. They have head and tail structure with dodecyl chain and phenylalkyl (benzyl or phenylethyl) groups. By introducing fluorine groups into the phenylalkyl groups, the electron transport properties were enhanced. In addition, modifying the substrate surfaces with a fluorine-rich polymer material and annealing at the appropriate temperature allowed the TFT devices with C12-NTCDI-F2Bn (R = 3,5-difluorobenzyl) to reach a maximum electron mobility (μ e) of 0.11 cm2 V−1 s−1. Furthermore, the alteration of the number of methylene carbons in the side chain of phenylalkyl groups in C12-NTCDI-R (namely, phenylethyl was used as R) affected the thermal behavior of their molecules during the thin-film formation. For the TFT device with C12-NTCDI-PhEt (R = phenylethyl), the μ e reached the maximum value of 0.16 cm2 V−1 s−1 by just surface modification, without annealing.

Pig eye area temperature extraction algorithm based on registered images

The body temperature of pigs is closely related to pig diseases. People have tried various methods to use infrared thermography technology for extracting pig body temperature. However, infrared images may vary with changes in pig body temperature. When the body temperature of pigs is abnormal, these body temperature extraction methods using only infrared images may not be able to locate the target area, thereby failing to complete the body temperature extraction operation. Taking the eye area as the target area, this study proposes apig eye area temperature extraction algorithmbased on registered images. This algorithm first aligns the infrared image with the visible image using registration techniques, then uses an object detection model to locate the pig's eye area on the visible light image, and finally completes the temperature extraction by combining the infrared temperature matrix. This algorithm uses visible images to locate the pig's eye area, fully utilizing the fact that visible images do not change with temperature, which can effectively avoid the problem that the pig's eye area may not be located by using only infrared images when the pig's body temperature is abnormal. To apply the pig eye detection model to pig farms and facilitate deployment on edge devices, this study conducts lightweight modification to the model by introducing the Ghost module to modify the neck and head structures of YOLOv7-tiny. The improved algorithm effectively reduces the number of parameters and computational cost. Compared with the original YOLOv7-tiny model, themean average precisiondecreased by 0.8%, while the number of parameters andcomputational costdecreased by 42.3% and 31.5%, respectively. The detection speed of the improved algorithm on the Jetson Nano (an edge device) increased by 12.8%. The improved algorithm achieves a balance betweendetection accuracyand detection speed, making it more suitable for deployment on edge devices.

An Introduction to the Comparison Shape of some Cranial Structures in Boleophthalmus Dussumieri and Periophthalmus Waltoni (Teleostei: Oxudercidae) in the Persian Gulf: An Exploratory Analysis with Micro-CT Scanning

Purpose: The osteological characteristics of fish, especially the head structure, are important in understanding the biological characteristics. Periophthalmus waltoni and Boleophthalmus dussumieri are among the mudskippers and are distributed along the coasts of the Oman Sea and the Persian Gulf.&#x0D; Materials and Methods: After catching and fixing the samples in 96% ethanol, the samples were then sent to the preclinical laboratory (Lotus-InVivo) for micro-CT scanning (TPCF, in Tehran university of medical sciences) for imaging.&#x0D; Results: In B. dussumieri, the skull is rudimentary and a high percentage of the bones is still cartilaginous. In this species, despite the larger head size, the braincase is small. In P. waltoni, the braincase is larger, but the skull tissue is completely bony and has very little cartilaginous. The jaws have also undergone drastic changes, corresponding to the change from a nearly fixed biting mouth to a flexible sucking mouth. In both species, the teeth are sharp and in two parts in the jaws. In P. waltoni, there are three pairs of sharp teeth for hunting in the upper jaw, the number of these teeth in B. dussumieri is four passes and it is less curved.&#x0D; Conclusion: In this report, for the first time, the skull structure of the Persian Gulf was investigated. Micro CT technique has also been used for the first time. Mudskippers have developed special adaptations to live in mud in terrestrial and aquatic conditions. These adaptations are greater in P. waltoni, which shows greater degrees of terrestrialization, and requires detailed studies in this field.

Conformational characteristics in Arabian and Thoroughbred horses

Horses are expected to maintain their racing lives with healthy and high performances. The body structure of an ideal race horse should have ideal athletic characteristics. By selecting horses with good body structure, it is possible to increase both race achievements and breeding values. Studies are ongoing to evaluate the ideal body structure of horses in an objective and measurable way. This helps identify the strengths and weaknesses of horses.&#x0D; The study was aimed at determining the overall body conformations of Arabian and Thoroughbred racing horses by morphometric measurements. Photographs of Arabian and Thoroughbred horses in standard positions were scaled on a computer to measure angle and length values. Using the data obtained, body structures were tried to be revealed for both races. In addition, structural differences that increased the risk of disability in both races were assessed. There were significant differences between the head, leg, and body structures of Arabian and Thoroughbred horses. Lover length measurements were found for Arabians than for Thoroughbreds. It was noted that differences between the two breed were significant, especially in distal extremity measures. It was found that in Thoroughbreds, the rump is generally higher than the withers, and this may result in a greater loading on the forelegs.&#x0D; In addition to identifying the body structures and differences of both breed, the results of the study are thought to be useful for selection practises and to contribute to the understanding of the aetiology of disabilities.

Aluminum Surface Defect Detection Algorithm Based on Improved YOLOv5

AbstractIndustrial defect detection is an important aspect of object detection. Aluminum is an indispensable material in the industrial field, but the complexity of surface defects on aluminum makes detection challenging. Therefore, the paper proposes the YOLOv5‐ESP algorithm based on the YOLOv5 algorithm. First, the problem of poor signal quality and small data samples is addressed through data enhancement. Second, the YOLOv5‐ESP incorporates the Efficient Channel Attention‐C3 (ECA‐C3) module in the Backbone structure to enhance attention toward defect regions. The Spatial Pooling Pyramid Cross‐Stage Partial Convolution (SPPCSPC) module is introduced to extract multi‐scale features of defects. The Poly‐scale Convolution (PSConv) is applied at the end of the Neck structure to resolve the problem of imprecise localization resulting from significant differences in the scale of defect features. Soft Non‐Maximum Suppression (Soft‐NMS) is utilized in the Head structure to optimize the prediction boxes. Experimental results show that the proposed YOLOv5‐ESP achieves a mean Average Precision (mAP) value of 94%, outperforming YOLOv5 and other classical algorithms. Furthermore, it maintains an average recognition time of 0.019 s per image, which meets the requirements of accuracy and real‐time performance in the industry.

Production of biomethane, biohydrogen, and volatile fatty acids from Nordic phytoplankton biomass grown in blended wastewater

Upgrading carbon-negative microalgal biomass to biofuels and value-added products presents a three-pronged solution for waste treatment, carbon capture, and economically viable bioenergy production. Acidogenesis and methanogenesis are versatile processes at the core of anaerobic digestion systems, facilitating the conversion of diverse biogenic substrates into energy and a wide range of biobased products. The present study was conducted to integrate acidogenesis and methanogenesis for coproduction of biohydrogen, biomethane, and volatile fatty acids from Nordic phytoplankton consortia. For this purpose, microalgal consortia was cultivated in a raceway pond equipped with high surface area structures. Harvested microalgal biomass was subjected to thermoalkaline (2% NaOH solution at 121 °C) and enzymatic (cellulase) pretreatments. The hydrolysates of the pretreated biomass were inoculated with thermally treated sludge for acidogenic fermentation and with mixture of untreated sludge and cow dung (1:1 v/v ratio) for anerobic digestion. The acidogenic process produced a significant amount of biohydrogen (maximum 164.8 mL bioH2/VSload) along with volatile fatty acids (maximum 7.9 g COD/L), while methanogenesis resulted in biomethane production of maximum 210.7 mL bioCH4/VSload, accompanied by an ammonium recovery of 1278 mg NH4+/L. These maximum yields were all achieved by enzymatic pretreatment of the biomass fraction harvested from high surface area brush head structures inserted in the raceway pond. These results have important implications for designing phytoplankton cultivation systems and upstream pathways to optimize energy production from carbon-negative phytoplankton biomass.

POPULATION STRUCTURE AND BARTONELLA QUINTANA IN HEAD AND BODY LICE IN POKHARA, NEPAL (ANOPLURA: PEDICULIDAE).

This study examined the population structure of head and body lice infesting a random sample of people in Pokhara, Nepal during 2003, 2004, and 2005. A total of 106 participants (6 to 72 yr old, median = 12) volunteered to have lice collected from their heads and clothing. Most participants (70%) harbored only head lice, some (15%) had only body lice, and some (15%) had concurrent infestations of head and body lice (dual infestations). A total of 1,472 lice was collected. Significantly more nymphs were collected than adult lice. Louse populations were generally small (geometric mean = 8.8 lice per person) but a few participants harbored larger louse populations (maximum = 65 lice). People with dual infestations harbored significantly more lice than people with single infestations; however, there was no difference in the infestation intensities between people infested with head lice only vs. those infested with body lice only. Male participants harbored significantly more lice than did females. There were no significant differences in infestation intensity due to participant age or their socioeconomic level. The sex ratio of adult lice was increasingly female biased with increasing adult louse density. Infection of lice with Bartonella quintana was low (ca. 1.5%). Pediculosis is a common problem in urban Nepal.

INNV-42. INTRA-ARTERIAL MELPHALAN FOR HISTIOCYTIC NEOPLASMS INVOLVING NEUROLOGIC, OCULAR, AND HEAD AND NECK STRUCTURES

Abstract BACKGROUND Histiocytic neoplasms are clonal hematopoietic disorders of myeloid lineage. Involvement of neurologic structures by histiocytic infiltration or disease-associated neurodegeneration (ND) represents a severe disease phenotype associated with morbidity and mortality. Despite recent advances with BRAF and MEK inhibitors, patients with neurohistiocytic involvement continue to demonstrate suboptimal outcomes. Intra-arterial chemotherapy (IAC) may represent an effective therapy for this disease manifestation. METHODS Patients with biopsy-proven histiocytic neoplasms involving neurologic, ocular, or head and neck structures were treated with intra-arterial melphalan delivered via selective catheterization of tumor arterial vasculature. Patients treated had refractory/persistent tumorous disease or progressive functional decline despite BRAF/MEK inhibition, or had limited disease appropriate for local therapy. Patients were evaluated for radiologic (PET or CT/MRI) response and functional neurologic response with quantified ataxia and dysarthria scales, or change in visual acuity (VA), as appropriate. Procedure-related adverse events as well as clinical and laboratory toxicities were captured. RESULTS 74 IAC treatments were administered to 17 patients, 12 with tumorous disease (6 Rosai-Dorfman disease; 3 xanthogranuloma; 3 Langerhans cell histiocytosis[LCH]; 1 Erdheim-Chester disease [ECD]), 5 with ND-LCH/ECD, and ten with disease-related impairment of VA. Patients received a range of 2-12 IAC doses. 9/12 (75%) patients with tumorous disease had complete or partial radiologic response. 4/5 patients with ND had measurable improvement in dysarthria and 2/5 had measurable improvement in ataxia. 4/10 patients had improved VA. Of all patients with clinical or radiologic response, three have experienced disease recurrence since IAC in long-term follow-up. No procedural adverse events were observed, 1 patient had neutropenia and 1 patient melphalan hypersensitivity. CONCLUSIONS IA melphalan is a safe and robustly effective treatment option for tumorous and ND histiocytic neoplasms of neurologic, ocular, and head and neck structures.

Anti-drone systems: An attention based improved YOLOv7 model for a real-time detection and identification of multi-airborne target

Recently, with the significant rise of drones, reinforcing and securing aerial security and privacy has become an urgent task. Their malicious use takes benefit from the malevolent deployment which leverages some existing gaps in Artificial Intelligence (AI) and cybersecurity. Anti-drone systems are the spotlighted security solution developed to ensure aerial safety and security against rogue drones. However, the anti-drone systems are constraints to accurate airborne target identification and real-time detection to neutralize the target properly without causing damages. In this paper, we have developed a real-time multi-target detection model based on Yolov7 aiming to detect, identify and locate the airborne target properly and rapidly using a varied dataset which is biased and imbalanced due to the differences between the targets. In order to develop a model with the best compromise between a high performance and fast speed, we have used a series of improvements by incorporating the CSPResNeXt module in the backbone, a transformer block with the C3TR attention mechanism and decoupled head structure to enhance the performance of the model. The comparative and ablation experiments confirm the effectiveness of the proposed ensemble learning-based model. The experiments have shown that the improved model has reached high performance, with 0.97 precision, 0.961 recall, 0.979 map@0.50 and 0.732 and 0.979 map@0.50–0.95. Additionally, the real-time detection condition is satisfied with 92 FPS and an inference speed equal to 0.02 ms per image. The results show that the model succeeds in achieving an optimal balance between inference speed and detection performance. The proposed model achieves competitive results compared with the existing state-of-the-art models.