Multiple Objects In Image Research Articles

Application of TensorFlow model for identification of herbaceous mimosa (Mimosa strigillosa) from digital images

This study explored the use of TensorFlow image recognition model to identify herbaceous mimosa (Mimosa strigillosa) from digital images. There is a demand for such technology toward digital mapping of the spatial distribution of these important perennial legumes in the context of pasture management and as well as management of reclamation ground cover landscapes. This study provided evidence of successful application of TensorFlow model for identification of herbaceous mimosa from digital images with final accuracy of 95 % or more. The complexity of ground images of multiple objects in this study is suspected to induce fluctuations in validation accuracy. Such fluctuation of the validation accuracy, however, was shown to decline over time as the accuracy increased with more processing epochs involved. Despite the downside of intensive data preparation and heavy computing resources, the approach tested in this study is promising toward the next step of the technology application for identification of herbaceous mimosa patches from images acquired using Unmanned Aerial Vehicle (UAV).

Three-dimensional imaging by compressed sensing based dual-frequency laser phase ranging

Scanless three-dimensional (3D) imaging technology has received extensive attention in recent years due to its rapid detection and system reliability. Compressed sensing imaging technology provides a new solution for the realization of scan-free 3D imaging. In this paper, a 3D imaging method based on dual-frequency laser phase ranging based on compressed sensing technology is introduced and realized. Using the combination of dual-frequency laser phase ranging and compressed sensing theory, two-dimensional range reconstruction from the time-domain light intensity signal collected by a single-point detector is performed. Aiming at the spatial sparsity of the target scene, this technology uses the compressed sensing algorithm to solve the phase information of the two-dimensional spatial distribution contained in the time domain signal so as to invert the 3D image information of the target scene and realize the effect of scanning-free 3D imaging. First, the feasibility of the system is verified by simulations, and the imaging effects of different reconstruction algorithms on different terrains are compared. Second, a non-scanning 3D imaging experimental platform is designed and built. Finally, the 3D images of multiple objects with 32 × 32 resolution are successfully reconstructed through experiments with a compression ratio of 0.25. The ranging accuracy of this system is 0.05 m. This work is promising for applications in multiple objects’ fast detections.

MVDet: multi-view multi-class object detection without ground plane assumption

Although many state-of-the-art methods of object detection in a single image have achieved great success in the last few years, they still suffer from the false positives in crowd scenes of the real-world applications like automatic checkout. In order to address the limitations of single-view object detection in complex scenes, we propose MVDet, an end-to-end learnable approach that can detect and re-identify multi-class objects in multiple images captured by multiple cameras (multi-view). Our approach is based on the premise that incorrect detection results in a specific view can be eliminated using precise cues from other views, given the availability of multi-view images. Unlike most existing multi-view detection algorithms, which assume that objects belong to a single class on the ground plane, our approach can classify multi-class objects without such assumptions and is thus more practical. To classify multi-class objects, we propose an integrated architecture for region proposal, re-identification, and classification. Additionally, we utilize the epipolar geometry constraint to devise a novel re-identification algorithm that does not require assumptions about ground plane assumption. Our model demonstrates competitive performance compared to several baselines on the challenging MessyTable dataset.

Amphiphilic branched polymer-nitroxides conjugate as a nanoscale agent for potential magnetic resonance imaging of multiple objects in vivo

BackgroundIn order to address the potential toxicity of metal-based magnetic resonance imaging (MRI) contrast agents (CAs), a concept of non-metallic MRI CAs has emerged. Currently, paramagnetic nitroxides (such as (2,2,5,5-tetramethylpyrrolidine-1-oxyl, PROXYL), (2,2,6,6-tetramethylpiperidine-1-oxide, TEMPO), etc.) are being extensively studied because their good stability and imaging mechanism are similar to metal-based contrast agents (such as Gd3+ chelate-based clinical CAs). However, a lower relaxivity and rapid in vivo metabolism of nitroxides remain to be addressed. Previous studies have demonstrated that the construction of macromolecular nitroxides contrast agents (mORCAs) is a promising solution through macromolecularization of nitroxides (i.e., use of large molecules to carry nitroxides). Macromolecular effects not only increase the stability of nitroxides by limiting their exposure to reductive substances in the body, but also improve the overall 1H water relaxation by increasing the concentration of nitroxides and slowing the molecular rotation speed.ResultsBranched pDHPMA-mPEG-Ppa-PROXYL with a high molecular weight (MW = 160 kDa) and a nitroxides content (0.059 mmol/g) can form a nanoscale (~ 28 nm) self-assembled aggregate in a water environment and hydrophobic PROXYL can be protected by a hydrophilic outer layer to obtain strong reduction resistance in vivo. Compared with a small molecular CA (3-Carboxy-PROXYL (3-CP)), Branched pDHPMA-mPEG-Ppa-PROXYL displays three prominent features: (1) its longitudinal relaxivity (0.50 mM− 1 s− 1) is about three times that of 3-CP (0.17 mM− 1 s− 1); (2) the blood retention time of nitroxides is significantly increased from a few minutes of 3-CP to 6 h; (3) it provides long-term and significant enhancement in MR imaging of the tumor, liver, kidney and cardiovascular system (heart and aortaventralis), and this is the first report on nitroxides-based MRI CAs for imaging the cardiovascular system.ConclusionsAs a safe and efficient candidate metal-free magnetic resonance contrast agent, Branched pDHPMA-mPEG-Ppa-PROXYL is expected to be used not only in imaging the tumor, liver and kidney, but also the cardiovascular system, which expands the application scope of these CAs.Graphical abstract

Channel and spatial attention based deep object co-segmentation

Object co-segmentation is a challenging task, which aims to segment common objects in multiple images at the same time. Generally, common information of the same object needs to be found to solve this problem. For various scenarios, common objects in different images only have the same semantic information. In this paper, we propose a deep object co-segmentation method based on channel and spatial attention, which combines the attention mechanism with a deep neural network to enhance the common semantic information. Siamese encoder and decoder structure are used for this task. Firstly, the encoder network is employed to extract low-level and high-level features of image pairs. Secondly, we introduce an improved attention mechanism in the channel and spatial domain to enhance the multi-level semantic features of common objects. Then, the decoder module accepts the enhanced feature maps and generates the masks of both images. Finally, we evaluate our approach on the commonly used datasets for the co-segmentation task. And the experimental results show that our approach achieves competitive performance.

Efficient Multi-Object Detection and Smart Navigation Using Artificial Intelligence for Visually Impaired People

Visually impaired people face numerous difficulties in their daily life, and technological interventions may assist them to meet these challenges. This paper proposes an artificial intelligence-based fully automatic assistive technology to recognize different objects, and auditory inputs are provided to the user in real time, which gives better understanding to the visually impaired person about their surroundings. A deep-learning model is trained with multiple images of objects that are highly relevant to the visually impaired person. Training images are augmented and manually annotated to bring more robustness to the trained model. In addition to computer vision-based techniques for object recognition, a distance-measuring sensor is integrated to make the device more comprehensive by recognizing obstacles while navigating from one place to another. The auditory information that is conveyed to the user after scene segmentation and obstacle identification is optimized to obtain more information in less time for faster processing of video frames. The average accuracy of this proposed method is 95.19% and 99.69% for object detection and recognition, respectively. The time complexity is low, allowing a user to perceive the surrounding scene in real time.

Object Detection through Modified YOLO Neural Network

In the field of object detection, recently, tremendous success is achieved, but still it is a very challenging task to detect and identify objects accurately with fast speed. Human beings can detect and recognize multiple objects in images or videos with ease regardless of the object’s appearance, but for computers it is challenging to identify and distinguish between things. In this paper, a modified YOLOv1 based neural network is proposed for object detection. The new neural network model has been improved in the following ways. Firstly, modification is made to the loss function of the YOLOv1 network. The improved model replaces the margin style with proportion style. Compared to the old loss function, the new is more flexible and more reasonable in optimizing the network error. Secondly, a spatial pyramid pooling layer is added; thirdly, an inception model with a convolution kernel of 1 ∗ 1 is added, which reduced the number of weight parameters of the layers. Extensive experiments on Pascal VOC datasets 2007/2012 showed that the proposed method achieved better performance.

The gold miner's dilemma: Use of information scent in cooperative and competitive information foraging

When searching for new information, do people focus their search on places not-yet discovered by others, or on places that others also focus on? Through a controlled experiment, we investigated heuristic rules that people adopt in social information search, a growing characteristic of how people find information in this hyperconnected world. Three people were connected online to simultaneously search for specific objects in multiple images, under either a cooperative or a competitive setting. They were provided with information about the current number of objects collected and the cumulative time spent on each image. People used such information to decide when to stop the current search and which image to explore next. Further, people paid more attention to others and distribute search efforts when cooperating, compared to when competing against others. Our findings highlight the heuristic rules that people adopt when searching in groups for new information.

Optimisation of linear dependence energy for object co‐segmentation in a set of images with heterogeneous contents

This work proposes a framework for simultaneously segmenting foreground objects in a collection of images having heterogeneous contents. Rather than resorting to image co-segmentation to segment similar objects in multiple images, which requires the use of categorised images, the authors’ idea disseminates segmentation information within images. In this way, it becomes easier to detect foreground objects in all of them simultaneously, mainly under the hypothesis of using similar or different images. General information is aggregated, on foregrounds as well as on backgrounds, from a set of images for joint segmentation of category-independent objects. The key idea is to estimate the linear dependence of the foreground histograms of the input images to optimise a Markov random field-based energy function. Iterative optimisation of each image permits after that the enhancement of the final segmentation results. Extensive experiments demonstrate that the proposed method (PM) enables full-object segmentation of foreground objects within a collection of images composed of different classes. Indeed, the validation of the accuracy on five challenging datasets (iCoseg, Oxford Flowers, MicroSoft Research Cambridge (MSRC), Caltech101 and Berkeley) shows that the PM achieves satisfactory results as compared with state-of-the-art methods. Besides, it has the challenging ability to efficiently deal with uncategorised objects.

Optimizing Infrared Camera Resolution for Small Object Detection using Subpixel Rendering and PIFS in Multiresolution Image Analysis.

Background: Breast cancer screening techniques have been developing rapidly in the field of imaging systems. One of these techniques is thermography, which is an alternative modality for mammography to detect breast lesions. Thermography utilization has been progressively developing as various models and methods of object processing improvement. Currently, the Fluke TIS20 infrared camera, with a resolution of 320 × 240, has not been used to measure precisely small objects such as early breast cancer lesions. Retrieval and processing of single images lead into imprecise object measurements and false positive results.Objective: Problems have been arisen due to the limitations of the camera resolution, object retrieval techniques and suboptimal image processing. The aim of this study was to detect accurately breast cancer lesions in rats, which were induced by carcinogenic compounds.Material and Methods: In this experimental study, development of models was conducted based on increasing image by optimizing the ability of low-resolution infrared (IR) cameras to identify s mall objects precisely. Image pixel density increased by adjusting the distance of the objects from the camera and multiple images of objects gradually shifting were used to measure object dimensions precisely.Results: The results showed that cancerous lesions as small as 1.27 mm could be detected. This method of lesion detection had a sensitivity and specificity of 93% and 77 % respectively.Conclusion: Small objects (cancerous lesions) were measured by increasing image resolution through splitting pixels into subpixels and combining several images using Partitioned Iterated Function Systems (PIFS).

Scene Description for Visually Impaired People with Multi-Label Convolutional SVM Networks

In this paper, we present a portable camera-based method for helping visually impaired (VI) people to recognize multiple objects in images. This method relies on a novel multi-label convolutional support vector machine (CSVM) network for coarse description of images. The core idea of CSVM is to use a set of linear SVMs as filter banks for feature map generation. During the training phase, the weights of the SVM filters are obtained using a forward-supervised learning strategy unlike the backpropagation algorithm used in standard convolutional neural networks (CNNs). To handle multi-label detection, we introduce a multi-branch CSVM architecture, where each branch will be used for detecting one object in the image. This architecture exploits the correlation between the objects present in the image by means of an opportune fusion mechanism of the intermediate outputs provided by the convolution layers of each branch. The high-level reasoning of the network is done through binary classification SVMs for predicting the presence/absence of objects in the image. The experiments obtained on two indoor datasets and one outdoor dataset acquired from a portable camera mounted on a lightweight shield worn by the user, and connected via a USB wire to a laptop processing unit are reported and discussed.

Experimental results of imaging objects in turbid liquid integrating multiview circularly polarized speckle images and deconvolution method

We hereby suggest and experimentally demonstrated an approach to enhance image quality during recovery of objects hidden in turbid liquid by fusion of single-shot multi-view circularly polarized speckle images collected by lens array and deconvolution algorithm based multiple medium sub-PSFs viewpoints. Circular polarized light was used since it has superior polarization memory enable deeper probing within turbid media. Experiments were carried out in transmission mode upon targets (transmitted and opaque) of different geometric shapes embedded within turbid liquid media of commercial instant coffee blended in lukewarm water at varied concentrations. The media were illuminated by circular polarized laser beam and multiple circular polarized speckled images were captured by camera. To counter the effect of scattering noises, multiple object images and Wiener deconvolution algorithm based multiple medium PSFs acquisition were shifted to a common center and then fused together to yield reconstructed images of the hidden objects. Following experiments, we demonstrated that by fusion of multiple images a better reconstruction was obtained as compared to the use of a single imaging lens. For a quantitative comparison, some common image quality metrics were used to evaluate the reconstruction quality. In addition, a phase only filter was applied during the deconvolution and its performance was compared with Wiener filtering. Overall, the feasibility of the proposed technique is proven by the experimental results. This work can be beneficial for application to imaging hidden targets buried in turbid system such as biological medium, fog, smoke, rain, etc. and therefore has potential for a wide variety of applications in medical diagnostics, security, sensing, underwater imaging, and more.

Electromagnetic Imaging of Dielectric Objects Using a Multi-Directional Search-Based Simulated Annealing.

In this paper, we introduce a global optimization method that is a novel combination of the simulated annealing method and the multi-directional search algorithm. We demonstrate the use of the algorithm for a microwave-imaging system to obtain the electrical properties of objects. The proposed global optimizer significantly improves the performance and speed of the simulated annealing method by utilizing a nonlinear simplex search, starting from an initial guess, and taking effective steps in obtaining the global solution of the minimization problem. Due to the efficient performance of the proposed global optimization method, we are able to obtain the shape, location, and material properties of the target without considering any a priori information about them. The accuracy and applicability of the proposed imaging method is demonstrated with some numerical results in which two-dimensional images of multiple objects are successfully reconstructed.

Elements marking of partially masked group objects using local descriptions of an associated continuous image

Questions of recognizing images of multiple objects whose dimensions are close to the resolution limit of recording equipment - groups of point objects (GPO) - are investigated. Images in the form of GPOs are found in radar images, whereas images of large-scale objects represented by characteristic points, clusters of data in a feature space, and events in mass service systems and automated control systems can be reduced to them. GPO recognition is complicated by disconnectedness of constituent elements, the narrowness of the image autocorrelation function in terms of the geometric transform parameters, spatial fluctuations of elements, false markings and omissions of signal markings. The most promising approaches to the image recognition of this type are based on the transformation of the GPO into a connected object - an associated continuous image (ACI) - and the analysis of its secondary features. However, for GPO with a non-stationary configuration and / or partially masked GPO, the recognition issues are still poorly investigated. The paper proposes an approach to the recognition of GPO based on the marking of their elements - the identification of point objects to within a metaclass (family, type) that is stable for nonstationary and partially masked GPOs. Four metaclasses are proposed: the extreme and inner points of the GPO with a chain structure, the outer and inner points of clusters. The essence of the approach to marking point marks is the recognition of complex-valued contour codes composed of samples of the cylindrical sections of an associated continuous image (ASI) around the GPO elements. To form the ASI, a vector field with partial sources in the elements of the GPO is used. Evaluations of the reliability of marking under various observation conditions are obtained. The results of marking are both of independent significance for various applied problems and suited for the subsequent recognition of GPO. Approaches to GPO recognition based on the results of marking of their elements are suggested.

Algorithm for extracting multiple object waves without Fourier transform from a single image recorded by spatial frequency-division multiplexing and its application to digital holography

We propose a novel algorithm that does not require any Fourier transform to extract multiple object waves in a single image recorded with spatial frequency-division multiplexing. Smoothing is utilized to extract the desired object-wave information from a spatially multiplexed image. Numerical and experimental results show its validity and applicability for image and Fresnel digital holography. Our investigations clarify the speeding up of both the object-wave extractions and multiple object–image reconstructions quantitatively.

Curating in the Postdigital Age

Curating in the Postdigital Age

Multi-view object detection in dual-energy X-ray images

Automatic inspection of X-ray scans at security checkpoints can improve the public security. X-ray images are different from photographic images. They are transparent. They contain much less texture. They may be highly cluttered. Objects may undergo in- and out-of-plane rotations. On the other hand, scale and illumination change is less of an issue. More importantly, X-ray imaging provides extra information which are usually not available in regular images: dual-energy imaging, which provides material information about the objects; and multi-view imaging, which provides multiple images of objects from different viewing angles. Such peculiarities of X-ray images should be leveraged for high-performance object recognition systems to be deployed on X-ray scanners. To this end, we first present an extensive evaluation of standard local features for object detection on a large X-ray image dataset in a structured learning framework. Then, we propose two dense sampling methods as keypoint detector for textureless objects and extend the SPIN color descriptor to utilize the material information. Finally, we propose a multi-view branch-and-bound search algorithm for multi-view object detection. Through extensive experiments on three object categories, we show that object detection performance on X-ray images improves substantially with the help of extended features and multiple views.

Nonrigid Registration Regularized by Shape Information: Application to Atlas Construction of Cardiac CT Images.

Cardiac atlases play an important role in the computer-aided diagnosis of cardiovascular diseases, in particular they need to deal with large and highly variable image datasets. In this paper, we propose a new nonrigid registration algorithm incorporating shape information, to produce comprehensive atlases. For one thing, the multiscale gradient orientation features of images are combined to form the construction of multifeature mutual information. Additionally, the shape information of multiple-objects in images is incorporated into the cost function for registration. We demonstrate the merits of the new registration algorithm on the 3D data sets of 15 patients. The experimental results show that the new registration algorithm can outperform the conventional intensity-based registration method. The obtained atlas can represent the cardiac structures more accurately.

An Innovative Multiple-Object Image Retrieval Framework Using Hierarchical Region Tree

Inaccurate image segmentation often has a negative impact on object-based image retrieval. Researchers have attempted to alleviate this problem by using hierarchical image representation. However, these attempts suffer from the inefficiency in building the hierarchical image representation and the high computational complexity in matching two hierarchically represented images. This paper presents an innovative multiple-object retrieval framework named Multiple-Object Image Retrieval (MOIR) on the basis of hierarchical image representation. This framework concurrently performs image segmentation and hierarchical tree construction, producing a hierarchical region tree to represent the image. In addition, an efficient hierarchical region tree matching algorithm is designed for multiple-object retrieval with a reasonably low time complexity. The experimental results demonstrate the efficacy and efficiency of the proposed approach.

Development of Correspondence Algorithm for Illumination Varying Dynamic Image Sequence using Symbolic Data Analysis: A Novel Approach

This paper presents a novel symbolic data analysis based correspondence algorithm to identify the correspondence of illumination varying multiple object image sequence for the dynamic image analysis. The correspondence is the matching of the similar data sets or entities or the features such as the region based global features, points, corners, edges between a set of images. The Correspondence of illumination varying multispectral, temporal multiple objects using symbolic data analysis of objects in the dynamic images is a highly challenging problem. The determining Correspondence of features is a high dimensional NP-hard problem, because of their exhaustive search for the features in the sequence of image frames. The current algorithm works on the basis of successive frame difference based segmentation. The minimum boundary rectangle constructed for segmented moving object region. Mean of the rectangle is considered as one of the numeric feature. Next the sequence converted to binary sequence based on the threshold and 8-connectivity of perimeter is another non numeric feature. Finally establish the correspondence using canonical symbolic data analysis. This work proposes the rectangular correspondence window, generated dynamically which depends on the size and the number of objects in the current frame. Here we have considered mean of minimum boundary rectangle and perimeter of the segmented region as the features. The experiment is conducted with single object in the scene and multiple objects in the scene of illumination varying, noisy image sequence. This experiment is carried out on the binary, gray and RGB color dynamic image sequence. This correspondence algorithm is especially suitable for tracking in indoor images illuminated by unconstrained multiple light sources, objects in the varying illumination environment, noisy image sequence, video surveillance and slow moving dynamic image sequence for indoor scenes of the constant background.