Medium-sized Objects Research Articles

Advanced Point Cloud Techniques for Improved 3D Object Detection: A Study on DBSCAN, Attention, and Downsampling

To address the challenges of limited detection precision and insufficient segmentation of small to medium-sized objects in dynamic and complex scenarios, such as the dense intermingling of pedestrians, vehicles, and various obstacles in urban environments, we propose an enhanced methodology. Firstly, we integrated a point cloud processing module utilizing the DBSCAN clustering algorithm to effectively segment and extract critical features from the point cloud data. Secondly, we introduced a fusion attention mechanism that significantly improves the network’s capability to capture both global and local features, thereby enhancing object detection performance in complex environments. Finally, we incorporated a CSPNet downsampling module, which substantially boosts the network’s overall performance and processing speed while reducing computational costs through advanced feature map segmentation and fusion techniques. The proposed method was evaluated using the KITTI dataset. Under moderate difficulty, the BEV mAP for detecting cars, pedestrians, and cyclists achieved 87.74%, 55.07%, and 67.78%, reflecting improvements of 1.64%, 5.84%, and 5.53% over PointPillars. For 3D mAP, the detection accuracy for cars, pedestrians, and cyclists reached 77.90%, 49.22%, and 62.10%, with improvements of 2.91%, 5.69%, and 3.03% compared to PointPillars.

MFLD: lightweight object detection with multi-receptive field and long-range dependency in remote sensing images

PurposeBased on the strong feature representation ability of the convolutional neural network (CNN), generous object detection methods in remote sensing (RS) have been proposed one after another. However, due to the large variation in scale and the omission of relevant relationships between objects, there are still great challenges for object detection in RS. Most object detection methods fail to take the difficulties of detecting small and medium-sized objects and global context into account. Moreover, inference time and lightness are also major pain points in the field of RS.Design/methodology/approachTo alleviate the aforementioned problems, this study proposes a novel method for object detection in RS, which is called lightweight object detection with a multi-receptive field and long-range dependency in RS images (MFLD). The multi-receptive field extraction (MRFE) and long-range dependency information extraction (LDIE) modules are put forward.FindingsTo concentrate on the variability of objects in RS, MRFE effectively expands the receptive field by a combination of atrous separable convolutions with different dilated rates. Considering the shortcomings of CNN in extracting global information, LDIE is designed to capture the relationships between objects. Extensive experiments over public datasets in RS images demonstrate that our MFLD method surpasses the state-of-the-art methods. Most of all, on the NWPU VHR-10 dataset, our MFLD method achieves 94.6% mean average precision with 4.08 M model volume.Originality/valueThis paper proposed a method called lightweight object detection with multi-receptive field and long-range dependency in RS images.

High-resolution topographic surveying and change detection with the iPhone LiDAR.

This paper introduces a comprehensive protocol leveraging open-access techniques to create small- to medium-scale 3D representations of the environment by using iPhone and iPad light detection and ranging (LiDAR). The protocol focuses on two capabilities of the iPhone LiDAR. The first capability is 3D modeling: iPhone LiDAR rapidly generates detailed indoor and outdoor 3D models, providing insights into object size, volume and geometry. The second capability is change detection: the 3D models created by the LiDAR sensor can be used for precise measurement of changes over time. Compared to other 3D topographic surveying methods, this method is rapid, high resolution, low cost and easy to use. The protocol outlines iPhone LiDAR scanning practices, model export and change detection. The expected results after executing the protocol are (i) a detailed 3D model of a small- to medium-sized object or area of interest and (ii) a distance point cloud revealing change between two point clouds of the same object or area between different times. The entire protocol can be conducted within 2 h by anyone with an iPhone with the LiDAR sensor and a computer. This protocol empowers scientists, students and community members conducting research with a cheap, easy-to-use method for addressing a range of questions and challenges, thus benefiting experts and the broader community.

Revolutionizing Depth Sensing: A Review Study of Apple LiDAR Sensor for as-built Scanning Applications

Incorporating the LiDAR sensor in the most recent Apple devices represents a substantial development in 3D mapping technology. Meanwhile, Apple's Lidar is still a new sensor. Therefore, this article reviews the potential uses of the Apple Lidar sensor in various fields, including engineering and construction, focusing on indoor and outdoor as-built 3D mapping and cultural heritage conservation. The affordable cost and shorter observation times compared to traditional surveying and other remote sensing techniques make the Apple Lidar an attractive choice among scholars and professionals. This article highlights the need for continued research on the Apple LiDAR sensor technology while discussing its specifications and limitations. A comprehensive review found that the Apple LiDAR sensor has shown promise in capturing 3D point clouds of small to medium-sized objects with exceptional detail. This technology offers a cost-effective and accessible option to scan areas faster and analyze data more quickly and automatically for 3D mapping and modelling in indoor and outdoor environments, particularly in areas with restricted access when using other traditional techniques. It also opens the door for more sophisticated applications in future studies, including cultural heritage conservation, archaeological investigations and feature detection, building health monitoring and many more.

TrashInst: real-time floating litter instance segmentation model

This paper introduces TrashInst, an innovative fully convolutional single-stage anchor-free real-time instance segmentation model designed specifically for the detection of floating litter in waterways. TrashInst features a streamlined Encoder-Decoder architecture and an efficient channel attention mask head, ensuring the preservation of critical feature maps while minimizing redundant computations. Even in scenarios with limited feature maps, our approach, complemented by sparse instance activation maps, excels in both detection and segmentation tasks. Our encoder effectively processes features derived from ResNet50 across multiple scales, generating comprehensive abstractions through the nested feature module. Subsequently, the Vortex Instance Activation Maps module (VIAM) aggregates these multi-scale nested features and integrates them with mask head outputs, yielding precise instance masks. Furthermore, we employ the focal Tversky objective function to balance the dataset. In terms of performance, our approach outperforms state-of-the-art real-time instance segmentation models, achieving a remarkable 34% accuracy improvement, all while maintaining swift execution at 44 frames per second (FPS) on a single NVIDIA GTX-3090 GPU. Rigorous validation underscores the delicate interplay between speed and precision, highlighting the model’s exceptional ability to handle objects of varying sizes in real-time settings. Significantly, TrashInst attains superior average precision (AP) results in detecting large to medium-sized objects, outperforming existing models by a notable 4∼22% in the large category and an impressive 1.1∼56% in the medium category within our dataset. For the benefit of the community data will be available at (https://github.com/nassim12/TrashInst).

An airport apron ground service surveillance algorithm based on improved YOLO network

<abstract> <p>To assure operational safety in the airport apron area and track the process of ground service, it is necessary to analyze key targets and their activities in the airport apron surveillance videos. This research shows an activity identification algorithm for ground service objects in an airport apron area and proposes an improved YOLOv5 algorithm to increase the precision of small object detection by introducing an SPD-Conv (spath-to-depth-Conv) block in YOLOv5's backbone layer. The improved algorithm can efficiently extract the information features of small-sized objects, medium-sized objects, and moving objects in large scenes, and it achieves effective detection of activities of ground service in the apron area. The experimental results show that the detection average precision of all objects is more than 90%, and the whole class mean average precision (mAP) is 98.7%. At the same time, the original model was converted to TensorRT and OpenVINO format models, which increased the inference efficiency of the GPU and CPU by 55.3 and 137.1%, respectively.</p> </abstract>

ASTROPHYSICS

Astrophysics is a branch of space science that applies the laws of physics and chemistry to seek to understand the universe and our place in it. The field explores topics such as the birth, life and death of stars, planets, galaxies, nebulae and other objects in the universe. The Astrophysics creates physical theories of small to medium-size objects and structures in the universe.

Development of a collaborative robotic system to polish leather surfaces

ABSTRACT Patina is the natural aging process that gives leather a unique, worn appearance. However, this patina effect can also be created artificially by using specific products to rub the leather surface. This research presents a new and innovative method to automate the process of creating patina on leather footwear using collaborative robots. These types of finishing tasks on medium size objects are delicate and often require adjustments by skilled craftsmen, adding complexity to the automation process. This new solution involves designing and implementing a collaborative tool specifically created to cultivate a patina finish on shoe surfaces. The system aims to speed up the process without compromising the quality of the finished product. In order to accomplish this task, a collaborative robot that can regulate contact force and a preexisting software feature for recording manual paths are utilized together with the skills of an expert craftsman. This enables the robot’s end effector to conform to the curved contours of the shoe. Furthermore, the tool’s path and position are modified based on the data from the CAD model to ensure precise outcomes. This study introduces a novel technique for automated and partially automated patina development in the shoe industry. This work presents a system integrated into a productive facility and it is currently being used.

A Vehicle Recognition Model Based on Improved YOLOv5

The rapid development of the automobile industry has made life easier for people, but traffic accidents have increased in frequency in recent years, making vehicle safety particularly important. This paper proposes an improved YOLOv5s algorithm for vehicle identification and detection to reduce vehicle driving safety issues based on this problem. In order to solve the problems of a disappearing model training gradient in the YOLOv5s algorithm, difficulty in recognizing small objects and poor recognition accuracy caused by the boundary frame regression function, it is necessary to implement a new function. These aspects have been enhanced in this article. On the basis of the traditional YOLOv5s algorithm, the ELU activation function is used to replace the original activation function. The attention mechanism module is then added to the YOLOv5s algorithm’s backbone network to improve the feature extraction of small and medium-sized objects. The CIoU Loss function replaces the original regression function of YOLOv5s, thereby enhancing the convergence rate and measurement precision of the loss function. In this paper, the constructed dataset is utilized to conduct pertinent experiments. The experimental results demonstrate that, compared to the previous algorithm, the mAP of the enhanced YOLOv5s is 3.1% higher, the convergence rate is 0.8% higher, and the loss is 2.5% lower.

A Methodical Approach to 3D Scanning of Heritage Objects Being under Continuous Display

Three dimensional digitization of cultural heritage resources gains a lot of attention from the European Union and the United Nations, which is clearly revealed in current strategic goals and financing perspectives. Existing methodological approaches to 3D scanning in a prevailing number of cases assume that the procedure of scanning is performed in places that are closed to tourists, at least for the time of scanning. However, closing an exhibition for tourists or moving an artifact to be scanned is not always possible. Thanks to the long-term experience of the authors with 3D scanning of cultural heritage, the special procedure was designed for small and medium size objects to overcome difficulties expected in such cases. The procedure has been successfully implemented during 3D scanning of objects exhibited in the Silk Road region (on the territory of modern Uzbekistan), as well as objects being parts of wooden sacral architecture of the Maramures region (in Romania). It was revealed that the proposed procedure was successfully allowed to counteract organizational problems during 3D scanning of heritage objects being under continuous display, and that the achieved results of scanning were nevertheless of good quality.

Sensing soluble molecules through SERS substrates in one-step procedure: unrevealing the Meiji woodblock printing materials

Ultrasensitive SERS substrates allowed us to detect complex mixtures of coloring components from Meiji Japanese woodblock prints (1868–1912). In museum settings, compositional analyses have limitations due to restrictions to sampling advised by conservators and curators for the adequate preservation of the objects. An additional layer of complexity is brought by the high heterogeneity of heritage materials, usually not resolved with commercial portable instruments. High-performance instruments for in situ analyses are seldom available in museums. Furthermore, the chambers of most instruments for morphological or chemical characterization accommodate small samples rather than large or medium-sized objects. The innovative sampling strategy herein proposed comprises the gentle touch-dry removal of small coloring molecules weakly bound to the surface of heritage objects, transferred through a silicone sampler to planar SERS substrates with selected solvents in a one-step procedure. The analytical protocol reduces the amount of sample necessary for reliable identification of color components down to nanograms. The selectivity of the solvents combined with the geometry of the planar SERS sensing devices produces reliable signals for molecular identification, with no need for incision or wetting of the printed material. Further, 3D Raman imaging allowed us to reach an unprecedented degree of molecular discrimination, advancing previously available minimally-invasive instrumental methods used in heritage science research. The validation with historical inks from Meiji woodblock prints led to the identification of soluble synthetic azo β-naphthols, barium sulfonic lakes, purple anilines, Prussian blue, glass arsenic sulfides and other traditional coloring media.

Visual Inspection Method for Metal Rolls Based on Multi-Scale Spatial Location Feature

Metal rolls in a non-ferrous-metal manufacturing workshop manifest the characteristics of symmetry, multiple scales and mutual covering, which poses great challenges for metal roll detection. To solve this problem, firstly, an efficient attention mechanism algorithm named ECLAM (efficient capture location attendant model) is proposed for capturing spatial position features efficiently, to obtain complete location information for metal rolls in a complex environment. ECLAM can improve the ability to extract the spatial features of backbone networks and reduce the influence of the non-critical background. In addition, in order to give feature maps a larger receptive field and improve the weight of location information in multi-scale feature maps, a nonlinear feature fusion module named LFFM (location feature fusion module) is used to fuse two adjacent feature images. Finally, a multi-scale object detection network named L-MSNet (location-based multi-scale object detection network) based on the combination of ECLAM and LFFM is proposed and used to accurately detect multi-scale metal rolls. In the experiments, multi-scale metal roll images are collected from an actual non-ferrous-metal manufacturing workshop. On this basis, a pixel-level image dataset is constructed. Comparative experiments show that, compared with other object detection methods, L-MSNet can detect multi-scale metal rolls more accurately. The average accuracy is improved by 2% to 5%, and the average accuracy of small and medium-sized objects is also significantly improved by 3% to 6%.

Assembly Design of Additive Manufacturing Products: A Computational Framework for Part Separation

Additive Manufacturing (AM) allows for the manufacturing of complex geometry parts as a single object regardless of their geometric complexity. However, manufacturing a consolidate object is not always to be better than the assembly approach. There are several practical scenarios which often require an assembly approach, i.e., (1) to achieve specific mechanical functionality in the product and aesthetic aspects; (2) to fit a product component in a predefined buildable space, particularly the case where large-scale products are required to be fitted with a developing platform space; and (3) to improve the additive manufacturing productivity (by minimizing processing time, improving product quality, surface finish and reducing material/support cost).In this paper, a computational framework is proposed to decompose the initial object design into several assemblies with the aim to minimize the total AM processing time. The decomposed components can be arranged into a minimum printing height space, thus it reduces the number of printing layer significantly, yielding to a build-up time reduction even if some time is spent in required assembling operations. Computation framework is implemented for three medium size object geometries. The presented study points out that the part decomposition and assembling approach can considerably reduce the total AM processing time. The methodology supports AM to be extended for a broad range of medium size products, and leads to strengthening the customizability of the products.

Retracted: KSSD: single‐stage multi‐object detection algorithm with higher accuracy

Considering that the single shot multibox detector (SSD) algorithm will be missed or even false when is used to detect the small- and medium-sized objects, in this study, Kullback–Leibler single shot multibox detection (KSSD) object detection algorithm is proposed to improve the accuracy of small- and medium-sized objects detection. Firstly, the details in the detection process are visualised with gradient-weighted class activation mapping technology, and the details of each detection layer are shown in the form of class activation maps. Then it is noted that the phenomenon of the false or missed detection of the objects to be detected on small- and medium-sized objects in the SSD algorithm is related to the regression loss function. Accordingly, Kullback–Leibler border regression loss strategy is adopted and non-maximum suppression algorithm is used to output the final prediction boxes. Experimental results show that compared with the existed detection algorithms, the improved algorithm in this study has higher accuracy and stability, and can significantly improve the detection effect on small- and medium-sized objects.

STUDIES ON EXTERNAL GEOMETRY OF A PLANE WITH PHOTOGRAMMETRIC METHODS AND LASER SCANNING

Abstract. The authors of the paper were looking for a simple methods of the examination the airframe structure framework as part of the scientific and research project. The main tasks were: carrying out tests and analyses in defining external geometry of a selected plane with three independent methods i.e., geodetic (surveying), photogrammetric and laser scanning and selection of one of these methods as fast and acceptable accurate (below 3 mm) measurement applied to examine plane geometry. The carried out experiments were mainly aimed at answering the question if tested measurement methods give satisfying and acceptable results in case of works referring to the objects such as plane and if the assumed accuracy is realistic to be achieved by these measurement techniques. The authors testing the available on the market methods of the measuring the geometry of planes in the aspect of the selection of most accurate, but at the same time, fast measurement method. Thus three independent methods were tested: classical geodetic measurement, laser scanning by means of two independent scanners and photogrammetric method (based on the generated cloud of points). Experiments carried out so far in various centres worldwide usually included measurements of single elements or small models of planes. In surveying measurements of large objects, mainly buildings are predominant (e.g. surveying or photogrammetry in the inventory of historic architectural monuments, studies of the geometry of engineering constructions, etc.). The objective of the experiment carried out by the authors was to find good solution somewhere in the middle, i.e., apply geodetic measurement methods, but apply them for so-called medium-size objects such as planes.

Remote-sensing hyperspectral imaging for applications in archaeological areas: Non-invasive investigations on wall paintings and on mural inscriptions in the Pompeii site

Hyper-Spectral Imaging (HSI) is nowadays a well-established technology in the field of Cultural Heritage (CH), being acknowledged as a highly effective tool for accomplishing both non-invasive diagnostics and documentation of different typologies of polychrome artworks. Until recently, in CH field HSI technique has been mainly applied for investigating polychrome surfaces of small or medium size objects in museum collections, such as easel paintings, illuminated manuscripts or paper based artifacts. Conversely, the outdoor applications of HSI on large mural surfaces, wall paintings, frescoes and archeological assets are still mostly unexplored. This is due to the fact that the HSI systems designed for CH applications usually work at short-distance from the targets and can cover only limited size surfaces. To overcome this limitation, and to extend the applications of HSI to investigating mural paintings in external contexts, cloisters, ceilings, or assets in archaeological sites, a remote-sensing approach, based on the use of readapted avionic sensors, has been considered and tested in-field.This work illustrates the results of the measurements campaigns carried out in the archaeological site of Pompeii (Italy) by means of remote-sensing HSI technology. To this purpose, the high-performance prototype SIMGA, an avionic hyperspectral imager developed by Leonardo Company for earth-surface observations, was employed for the first time in an archeological context, within the framework of the wider project “Great Pompeii Project”. HSI measurements were performed on different types of wall surfaces, including mural paintings and mural inscriptions. On the wall paintings, the acquired HSI data proved to be effectively usable for identification of most of pictorial materials and for mapping their distributions. The presence of gypsum as marker of degradation phenomena could be detected and mapped. On mural inscriptions, the HSI data could be suitably processed to enhance some vanishing traits, hardly perceivable by visual inspection. The proposed approach provided evidence that HSI is a promising tool to retrieve some faded parts of partially lost texts of mural inscriptions in archeological contexts.

Thinking Ecologically, Knowing Responsibly

This essay extends my engagements with questions of epistemic agency and the politics of epistemic location, in Epistemic Responsibility and in Ecological Thinking to consider how questions of understanding and of certainty play diversely into human and other ecological circumstances. In so doing, it opens lines of inquiry not immediately available in standard western-northern approaches to epistemology with their concentration on medium-sized physical objects in their presupposed neutrality and replicability. Working from a tacit assumption that knowing and knowers are always situated, and that they are enabled or restricted in so being, the book engages with specific epistemic situations in order to show how “situatedness” indeed makes knowledge possible, while regarding it as an enabling rather than a constraining modality.

Biochemical sensor simulation for rectangular metal nanoparticles

To enhance sensor sensitivity and resolution, the present study proposes a biochemical sensor utilizing localized surface plasmon resonance to detect minute changes in the refractive index of the analyte. This sensor comprises a single-mode fiber located on the front and back ends of the sensor, which is responsible for the input and output of fiber optic signals, and two geometrical objects arranged in an interwoven pattern in the sensing region at the center of the sensor. In particular, the rectangular metal nanoparticles are arranged in a rectangular fashion and plated in the etched cladding layer to increase the number of nanosized metal particles on the X–Y plane. In addition, to decrease the calculation time and memory capacity while increasing the precision of simulations, a numerical calculation approach with power functions was employed. This approach comprises the finite element method, eigenmode expansion method, perfectly matched layer, perfectly reflecting boundary condition, boundary meshing method, and object meshing method. Triangular meshes were employed as the basis and sensors were divided into object boundaries, small-sized objects, medium-sized objects, and large-sized objects. The meshing ratio was set to 1:9:181:1810. Numerical simulation results indicate that the proposed localized surface plasmon resonance sensor achieved superior length (approx. 170 μm), resolution (approx. −155 dB), and sensitivity (approx. 137,664.4 nm/RIU).

X-Ray Computed Tomography In Situ: An Opportunity for Museums and Restoration Laboratories

X-ray Computed Tomography (X-ray CT) is a sophisticated non-destructive imaging technique to investigate structures and materials of complex objects, and its application can answer many conservation and restoration questions. However, for Cultural Heritage investigations, medical CT scanners are not optimized for many case-studies: These instruments are designed for the human body, are not flexible and are difficult to use in situ. To overcome these limitations and to safely investigate works of art on site—in a restoration laboratory or in a museum—the X-ray Tomography Laboratory of the University of Bologna designed several CT systems. Here we present two of these facilities and the results of important measurement campaigns performed in situ. The first instrument, light and flexible, is designed to investigate medium-size objects with a resolution of a few tens of microns and was used for the CT analysis of several Japanese theater masks belonging to the collection of the “L. Pigorini” Museum (Rome). The second is designed to analyze larger objects, up to 200 cm and was used to investigate the collection of the so-called “Statue Vestite” (devotional dressed statues) of the Diocesan Museum of Massa.

Broadband extended source imaging Mueller-matrix polarimeter.

An imaging Mueller matrix polarimeter, named the red-green-blue (RGB)950, takes images of medium-sized (tens of centimeters) objects by using a very bright source, large polarization state generator, and high-quality camera. Its broadband extended light source switches between red, green, blue, and near-infrared light to allow taking polarimetric images for comparison with RGB camera images. The large diffuse source makes shadow transitions gradual and spreads out the specular reflected spot into a larger less conspicuous area.