Imaging Multi-Object Research Articles

Swarm intelligence for full Stokes dynamic imaging reconstruction of interferometric data

Context. In very long baseline interferometry (VLBI), the combination of multiple antennas permits the synthesis of a virtual telescope with a larger diameter and consequently higher resolution than the individual antennas. However, due to the sparse nature of the array, recovering an image from the observed data is a challenging ill-posed inverse problem. Aims. The VLBI community is interested in not only recovering an image in total intensity from interferometric data, but also in obtaining results in the polarimetric and the temporal domain. Only a few algorithms are able to work in all these domains simultaneously. In particular, the algorithms based on optimization that consider various penalty terms specific to static total intensity imaging, time-variability and polarimetry are restricted to grids in the domain of the objective function. In this work we present a novel algorithm, multiobjective particle swarm optimization (MO-PSO), that is able to recover the optimal weights without any space-gridding, and to obtain the marginal contribution of each of the playing terms. Methods. To this end, we utilized multiobjective optimization together with particle swarm metaheuristics. We let the swarm of weights converge to the best position. Results. We evaluate our algorithm with synthetic data sets that are representative for the main science targets and instrumental configuration of the Event Horizon Telescope Collaboration (EHTC) and its planned successors. We successfully recover the polarimetric, static, and time-dynamic signature of the ground truth movie' even with relative sparsity, and a set of realistic data corruptions. Conclusions. We have built a novel, fast, hyperparameter space gridding-free algorithm that successfully recovers static and dynamic polarimetric reconstructions. Compared to regularized maximum likelihood (RML) methods, it avoids the need for parameter surveys, and it is not limited to the number of pixels, unlike recently proposed multiobjective imaging algorithms. Hence, this technique is a novel useful alternative tool to characterize full Stokes time-(in)dependent signatures in a VLBI data set robustly with a minimal set of user-based choices.

Multi-Objective Region Encryption Algorithm Based on Adaptive Mechanism

The advancement of information technology has led to the widespread application of remote measurement systems, where information in the form of images or videos, serving as measurement results, is transmitted over networks. However, this transmission is highly susceptible to attacks, tampering, and disputes, posing significant risks to the trustworthy transmission of measurement results from instruments and devices. In recent years, many encryption algorithms proposed for images have focused on encrypting the entire image, resulting in resource waste. Additionally, most encryption algorithms are designed only for single-object-type images. Addressing these issues, this paper proposes a multi-object region encryption algorithm based on an adaptive mechanism. Firstly, an adaptive mechanism is employed to determine the strategy for adjusting the sampling rate of encryption objects, achieved through an encryption resource allocation algorithm. Secondly, an improved polygon segmentation algorithm is utilized to separate single-object regions from multi-object images, dynamically adjusting the sequence of encryption objects based on the adaptive mechanism. Finally, encryption is achieved using a chaos fusion XOR encryption algorithm. Experimental validation using instrument images demonstrates that the proposed algorithm offers high efficiency and security advantages compared to other mainstream image encryption algorithms.

Super-agile satellites imaging mission planning method considering degradation of image MTF in dynamic imaging

Super-agile satellites with dynamic imaging capabilities can significantly improve the imaging efficiency of space-borne Earth observation. However, the image modulation transfer function (image MTF) of dynamic imaging varies dramatically with changes in the attitude maneuver status while improving observation efficiency. Therefore, both the observation efficiency and image MTF should be considered in the imaging mission planning of super-agile satellites. Existing imaging mission planning methods are mainly aimed at traditional agile satellites that are not capable of dynamic imaging, and the mainly optimization objectives are coverage revenue, spatial resolution, and task performance time, without considering the observation efficiency and image MTF as the optimization objectives. To overcome the problem of image MTF degradation caused by dynamic imaging, a mission planning method which considers the degradation of the image MTF and observation efficiency was proposed. Firstly, a mapping relationship between the angular velocity of the attitude maneuvering and the image MTF was constructed. On this basis, a Pareto multi-objective imaging mission planning model of partial order based was established. Considering the particularity of the model, the non-dominated sorting strategy of the NSGA-II algorithm was redesigned when solving the model using NSGA-II algorithm. Finally, the imaging tasks of different scales in single satellite and multi-satellite simulation scenes were set, and the proposed imaging mission planning method was verified. The planning results showed that the proposed method can achieve a balanced optimization of the image MTF and task efficiency under the premise of ensuing maximum observation coverage revenue.

Optimizing Image Enhancement: Feature Engineering for Improved Classification in AI-Assisted Artificial Retinas.

Artificial retinas have revolutionized the lives of many blind people by enabling their ability to perceive vision via an implanted chip. Despite significant advancements, there are some limitations that cannot be ignored. Presenting all objects captured in a scene makes their identification difficult. Addressing this limitation is necessary because the artificial retina can utilize a very limited number of pixels to represent vision information. This problem in a multi-object scenario can be mitigated by enhancing images such that only the major objects are considered to be shown in vision. Although simple techniques like edge detection are used, they fall short in representing identifiable objects in complex scenarios, suggesting the idea of integrating primary object edges. To support this idea, the proposed classification model aims at identifying the primary objects based on a suggested set of selective features. The proposed classification model can then be equipped into the artificial retina system for filtering multiple primary objects to enhance vision. The suitability of handling multi-objects enables the system to cope with real-world complex scenarios. The proposed classification model is based on a multi-label deep neural network, specifically designed to leverage from the selective feature set. Initially, the enhanced images proposed in this research are compared with the ones that utilize an edge detection technique for single, dual, and multi-object images. These enhancements are also verified through an intensity profile analysis. Subsequently, the proposed classification model's performance is evaluated to show the significance of utilizing the suggested features. This includes evaluating the model's ability to correctly classify the top five, four, three, two, and one object(s), with respective accuracies of up to 84.8%, 85.2%, 86.8%, 91.8%, and 96.4%. Several comparisons such as training/validation loss and accuracies, precision, recall, specificity, and area under a curve indicate reliable results. Based on the overall evaluation of this study, it is concluded that using the suggested set of selective features not only improves the classification model's performance, but aligns with the specific problem to address the challenge of correctly identifying objects in multi-object scenarios. Therefore, the proposed classification model designed on the basis of selective features is considered to be a very useful tool in supporting the idea of optimizing image enhancement.

Intelligent system for solid waste classification using combination of image processing and machine learning models

ABSTRACT Solid waste is a major issue in all cities around the world. Classification and segregation of solid waste prior to reuse, recycle or recover is an important step towards sustainable waste management. Traditional manual sorting of solid waste is a labour-intensive process that may pose health risks to the workers. Currently, automated classification of solid waste using machine learning techniques is widely applied. This study aims to develop an automated waste classification model by testing traditional and deep machine learning models. To achieve that, both open and generated datasets were used in the model training and testing. The study results showed relatively low prediction capability of the traditional machine learning models like Random Forest (RF) and Support Vector Machine (SVM) as compared to the deep machine learning Convolutional Neural Network (CNN). The testing of the three models on a combined data set of Trashnet with local garbage data set resulted in accuracy of 62.5% for SVM, 72.0% for RF and 92.7% for CNN. JONET deep learning model has been developed using a combination of pre-trained base model (DenseNet 201) with a new architicture that contains a fully connected layer in the classification stage with 1024 neurons. The model is capable to identify six classes of solid waste items with various accuracies. When tested on the Trashnet dataset, the accuracy was 96.06%, while testing on the local garbage dataset gave an accuracy of 94.40%. JONET has been tested also on multi object images which gave an acceptable prediction accuracy.

The distinguish between cats and dogs based on Detectron2 for automatic feeding

With the rapid growth of urbanization, the problem of stray animals on the streets is particularly prominent, especially the shortage of food for cats and dogs. This study introduces an automatic feeding system based on the Detectron2 deep learning framework, aiming to accurately identify and provide suitable food for these stray animals. Through training using Detectron2 with a large amount of image data, the system shows extremely high recognition accuracy in single-object images. When dealing with multi-object images, Detectron2 can generate independent recognition frames for each target and make corresponding feeding decisions. Despite the outstanding performance of the model, its potential uncertainties and errors still need to be considered. This research not only offers a practical solution to meet the basic needs of stray animals but also provides a new perspective for urban management and animal welfare. By combining technology with social responsibility, this innovative solution opens up a new path for solving the stray animal problem in cities, with broad application prospects and profound social significance.

Prior-free mixed speckle simplex separation strategy for multi-object imaging through thin scattering media beyond the optical memory effect

The optical technique of imaging through scattering media based on the optical memory effect (OME) sustains a limited field-of-view (FOV). Therefore, a prior-free imaging method is proposed to reconstruct multiple objects through the scattering media beyond the OME range. Based on the mixed speckle simplex separation strategy designed in this method, separating speckles of sub-objects is simplified as seeking the vertices of the mixed speckle simplex. An effective initial analysis of the mixed speckle simplex constructed by random intensity modulation is provided by vertex component analysis. The exact speckles of sub-objects are then separated by the specially designed non-negative matrix factorization algorithm. The multiple hidden objects can be recovered from the separated speckles respectively. The feasibility and imaging effects of the proposed method have been demonstrated via experiments. Multi-object imaging through the scattering media beyond at least four times the OME range has been realized. This work effectively advances speckle separation strategies to enlarge the limited FOV for imaging through scattering media.

Applying Object Detection and Embedding Techniques to One-Shot Class-Incremental Multi-Label Image Classification

In this paper, we introduce an efficient approach to multi-label image classification that is particularly suited for scenarios requiring rapid adaptation to new classes with minimal training data. Unlike conventional methods that rely solely on neural networks trained on known classes, our model integrates object detection and embedding techniques to allow for the fast and accurate classification of novel classes based on as few as one sample image. During training, we use either Convolutional Neural Network (CNN)- or Vision Transformer-based algorithms to convert the provided sample images of new classes into feature vectors. At inference, a multi-object image is analyzed using low-threshold object detection algorithms, such as YOLOS or CutLER, identifying virtually all object-containing regions. These regions are subsequently converted into candidate vectors using embedding techniques. The k-nearest neighbors are identified for each candidate vector, and labels are assigned accordingly. Our empirical evaluation, using custom multi-label datasets featuring random objects and backgrounds, reveals that our approach substantially outperforms traditional methods lacking object detection. Notably, unsupervised object detection exhibited higher speed and accuracy than its supervised counterpart. Furthermore, lightweight CNN-based embeddings were found to be both faster and more accurate than Vision Transformer-based methods. Our approach holds significant promise for applications where classes are either rarely represented or continuously evolving.

A Deep Recurrent Learning-Based Region-Focused Feature Detection for Enhanced Target Detection in Multi-Object Media.

Target detection in high-contrast, multi-object images and movies is challenging. This difficulty results from different areas and objects/people having varying pixel distributions, contrast, and intensity properties. This work introduces a new region-focused feature detection (RFD) method to tackle this problem and improve target detection accuracy. The RFD method divides the input image into several smaller ones so that as much of the image as possible is processed. Each of these zones has its own contrast and intensity attributes computed. Deep recurrent learning is then used to iteratively extract these features using a similarity measure from training inputs corresponding to various regions. The target can be located by combining features from many locations that overlap. The recognized target is compared to the inputs used during training, with the help of contrast and intensity attributes, to increase accuracy. The feature distribution across regions is also used for repeated training of the learning paradigm. This method efficiently lowers false rates during region selection and pattern matching with numerous extraction instances. Therefore, the suggested method provides greater accuracy by singling out distinct regions and filtering out misleading rate-generating features. The accuracy, similarity index, false rate, extraction ratio, processing time, and others are used to assess the effectiveness of the proposed approach. The proposed RFD improves the similarity index by 10.69%, extraction ratio by 9.04%, and precision by 13.27%. The false rate and processing time are reduced by 7.78% and 9.19%, respectively.

Large-Scale Multi-Objective Imaging Satellite Task Planning Algorithm for Vast Area Mapping

With satellite quantity and quality development in recent years, remote sensing products in vast areas are becoming widely used in more and more fields. The acquisition of large regional images requires the scientific and efficient utilization of satellite resources through imaging satellite task planning technology. However, for imaging satellite task planning in a vast area, a large number of decision variables are introduced into the imaging satellite task planning model, making it difficult for existing optimization algorithms to obtain reliable solutions. This is because the search space of the solution increases the exponential growth with the increase in the number of decision variables, which causes the search performance of optimization algorithms to decrease significantly. This paper proposes a large-scale multi-objective optimization algorithm based on efficient competition learning and improved non-dominated sorting (ECL-INS-LMOA) to efficiently obtain satellite imaging schemes for large areas. ECL-INS-LMOA adopted the idea of two-stage evolution to meet the different needs in different evolutionary stages. In the early stage, the proposed efficient competitive learning particle update strategy (ECLUS) and the improved NSGA-II were run alternately. In the later stage, only the improved NSGA-II was run. The proposed ECLUS guarantees the rapid convergence of ECL-INS-LMOA in the early evolution by accelerating particle update, introducing flight time, and proposing a binary competitive swarm optimizer BCSO. The results of the simulation imaging experiments on five large areas with different scales of decision variables show that ECL-INS-LMOA can always obtain the imaging satellite mission planning scheme with the highest regional coverage and the lowest satellite resource consumption within the limited evaluation times. The experiments verify the excellent performance of ECL-INS-LMOA in solving vast area mapping planning problems.

Comparison of Terahertz Transceiver Array Topologies in Multiobject Imaging

We investigate the effect of side lobe level (SLL) and multipath propagation in millimeter-wave(mmWave) and terahertz (THz) imaging systems for various transceiver array topologies. This work considers practical limitations in high-resolution active imaging systems and focuses on electrically large, planar antenna array topologies. As such, we investigate the imaging performance of sparse transmitting/receiving (Tx/Rx) array topologies that enable high angular/spatial resolution imaging with minimum hardware complexity. We show that 1) properly tapering the sparse array excitation can mitigate image clutter caused by high side lobes and 2) sparse arrays are more vulnerable to multipath reflections and clutter in dense scenes with highly reflective surfaces. We present both full-wave numerical analysis and laboratory experiments (250-300 GHz) to verify and compare the imaging array performance for several exemplary scenes.

Tea leaf disease detection and identification based on YOLOv7 (YOLO-T)

A reliable and accurate diagnosis and identification system is required to prevent and manage tea leaf diseases. Tea leaf diseases are detected manually, increasing time and affecting yield quality and productivity. This study aims to present an artificial intelligence-based solution to the problem of tea leaf disease detection by training the fastest single-stage object detection model, YOLOv7, on the diseased tea leaf dataset collected from four prominent tea gardens in Bangladesh. 4000 digital images of five types of leaf diseases are collected from these tea gardens, generating a manually annotated, data-augmented leaf disease image dataset. This study incorporates data augmentation approaches to solve the issue of insufficient sample sizes. The detection and identification results for the YOLOv7 approach are validated by prominent statistical metrics like detection accuracy, precision, recall, mAP value, and F1-score, which resulted in 97.3%, 96.7%, 96.4%, 98.2%, and 0.965, respectively. Experimental results demonstrate that YOLOv7 for tea leaf diseases in natural scene images is superior to existing target detection and identification networks, including CNN, Deep CNN, DNN, AX-Retina Net, improved DCNN, YOLOv5, and Multi-objective image segmentation. Hence, this study is expected to minimize the workload of entomologists and aid in the rapid identification and detection of tea leaf diseases, thus minimizing economic losses.

A Two-limb Explanation for the Optical-to-infrared Transmission Spectrum of the Hot Jupiter HAT-P-32Ab

We present a new optical transmission spectrum of the hot Jupiter HAT-P-32Ab acquired with the Carnegie Observatories Spectrograph and Multiobject Imaging Camera (COSMIC) on the Palomar 200 inch Hale Telescope (P200). The P200/COSMIC transmission spectrum, covering a wavelength range of 3990–9390 Å, is composed of 25 spectrophotometric bins with widths ranging from 200 to 400 Å and consistent with previous transit measurements obtained in the common wavelength range. We derive a combined optical transmission spectrum based on measurements from five independent instruments, which, along with the 1.1–1.7 μm spectrum acquired by the Hubble Space Telescope and two Spitzer measurements, exhibits an enhanced scattering slope blueward of a relatively flat optical continuum, a water absorption feature at 1.4 μm, and a carbon dioxide feature at 4.4 μm. We perform Bayesian spectral retrieval analyses on the 0.3–5.1 μm transmission spectrum and find that it can be well explained by a two-limb approximation of solar metallicity, with a strongly hazy morning limb of K and a haze-free evening limb of K. This makes HAT-P-32Ab a promising target for James Webb Space Telescope to look for asymmetric signatures directly in the light curves.

Multi-Objective Image Optimization of Product Appearance Based on Improved NSGA-Ⅱ

Multi-Objective Image Optimization of Product Appearance Based on Improved NSGA-Ⅱ

ProMIS: probability-based multi-object image synthesis-relevant weakly supervised object detection method

ProMIS: probability-based multi-object image synthesis-relevant weakly supervised object detection method

Cat Breeds Classification Using Convolutional Neural Network For Multi-Object Image

Cat is one of the most popular pets. There are many cat breeds with unique characteristic and treatment for each breed. A cat owner can have more than one cat, either the same breed or different breeds. But not all cat owners know the breeds of their cats. Computers can be trained to recognized cat breeds, but there are many challenges for computers because it limited by how much they have been trained and programmed. In recent years, a lot of research about image classification has been done before and got various result, but most of the data used in previous research were single object images. Therefore, this study of cat breeds classification would be conducted with Convolutional Neural Network (CNN) in the Multi-Object images. This method was chosen because it had good classification results in the previous studies. This study used 5 breeds of cats with every breed having 200-3200 images for training. The test results were measured using confusion matrix, obtaining the precision, recall, f1 score and accuracy of 100% on multi-object images with 2 objects and 3 objects. On images with 4 objects achieved the precision, recall, f1 score and accuracy value of 89%, 87%, 87% and 95%. While the value of precision, recall, f1 score and accuracy on images with 5 objects get 87%, 86%, 86% and 94%, respectively.

Automated bridge surface crack detection and segmentation using computer vision-based deep learning model

Bridge maintenance will become a widespread trend in the engineering industry as the number of bridges grows and time passes. Cracking is a common problem in bridges with concrete structures. Allowing it to expand will result in significant economic losses and accident risks This paper proposed an automatic detection and segmentation method of bridge surface cracks based on computer vision deep learning models. First, a bridge surface crack detection and segmentation dataset was established. Then, according to the characteristics of the bridge, we improved the You Only Look Once (YOLO) algorithm for bridge surface crack detection. The improved algorithm was defined as CR-YOLO, which can identify cracks and their approximate locations from multi-object images. Subsequently, the PSPNet algorithm was improved to segment the bridge cracks from the non-crack regions to avoid the visual interference of the detection algorithm. Finally, we deployed the proposed bridge crack detection and segmentation algorithm in an edge device. The experimental results show that our method outperforms other baseline methods in generic evaluation metrics and has advantages in Model Size(MS) and Frame Per Second (FPS).

EdgeXAR: A 6-DoF Camera Multi-target Interaction Framework for MAR with User-friendly Latency Compensation

The computational capabilities of recent mobile devices enable the processing of natural features for Augmented Reality (AR), but the scalability is still limited by the devices' computation power and available resources. In this paper, we propose EdgeXAR, a mobile AR framework that utilizes the advantages of edge computing through task offloading to support flexible camera-based AR interaction. We propose a hybrid tracking system for mobile devices that provides lightweight tracking with 6 Degrees of Freedom and hides the offloading latency from users' perception. A practical, reliable and unreliable communication mechanism is used to achieve fast response and consistency of crucial information. We also propose a multi-object image retrieval pipeline that executes fast and accurate image recognition tasks on the cloud and edge servers. Extensive experiments are carried out to evaluate the performance of EdgeXAR by building mobile AR apps upon it. Regarding the Quality of Experience (QoE), the mobile AR apps powered by EdgeXAR framework run on average at the speed of 30 frames per second with precise tracking of only 1-2 pixel errors and accurate image recognition of at least 97% accuracy. As compared to Vuforia, one of the leading commercial AR frameworks, EdgeXAR transmits 87% less data while providing a stable 30FPS performance and reducing the offloading latency by 50 to 70% depending on the transmission medium. Our work facilitates the large-scale deployment of AR as the next generation of ubiquitous interfaces.

A Novel U-Net Based Deep Learning Method for 3D Cardiovascular MRI Segmentation.

Medical multiobjective image segmentation aims to group pixels to form multiple regions based on the different properties of the medical images. Segmenting the 3D cardiovascular magnetic resonance (CMR) images is still a challenging task owing to several reasons, including individual differences in heart shapes, varying signal intensities, and differences in data signal-to-noise ratios. This paper proposes a novel and efficient U-Net-based 3D sparse convolutional network named SparseVoxNet. In this network, there are direct connections between any two layers with the same feature-map size, and the number of connections is reduced. Therefore, the SparseVoxNet can effectively cope with the optimization problem of gradients vanishing when training a 3D deep neural network model on small sample data by significantly decreasing the network depth, and achieveing better feature representation using a spatial self-attention mechanism finally. The proposed method in this paper has been thoroughly evaluated on the HVSMR 2016 dataset. Compared with other methods, the method achieves better performance.

Defocus diagnosis method of LAMOST spectrograph based on multi-object image definition evaluation

Defocus diagnosis method of LAMOST spectrograph based on multi-object image definition evaluation