Object Localization In Images Research Articles

An approximate method of estimating the depth of objects using wavelet-based transformation

Modern three-dimensional computer vision technologies are actively developing, offering new solutions based on the analysis of environmental details. One of the key tasks is to determine the depth of location of objects in images, which requires efficient and fast methods of their processing. This study proposes a new depth estimation method for 3D computer vision based on wavelet transform optimization. Solutions are offered to simplify the disparity calculation, which is traditionally used for the construction of depth maps, with the possibility of describing contours with adjustable detail based on the wavelet transform. The main advantage of this method is the increase in performance due to the allocation of the Haar wavelet carrier length in the extremum search area. The simulation confirmed the effectiveness of the proposed approach for constructing depth maps, which allows us to recommend it for use in unmanned aerial vehicles (UAVs) in conditions of limited computing and energy resources.

Towards reliable object representation via sparse directional patches and spatial center cues

In the process of image understanding, the human visual system (HVS) performs multiscale analysis on various objects. HVS primarily focuses on marginally conspicuous image patches located within or around distinct objects rather than scanning the image pixels point by point. Inspired by the HVS mechanism, in this paper, we aimed to describe and exploit multiscale decomposition-based patch detection models for automatic visual feature representation and object localization in images. Our investigation into mimicking and modeling the HVS to capture conspicuous sparse patches and their spatial distribution clues makes a profound contribution to the automatic comprehension and characterization of images by machines. This study demonstrates that the sparse patch-based visual representation with spatial center cues is intrinsically tolerant to object positioning and understanding beyond object variations in spatial position, multiresolution, and chrominance, which has significant implications for many vision-based automatic object grabbing and perception applications, such as robotics, human‒machine interaction, and unmanned aerial vehicles (UAVs).

P-243 Automating tracking of cell division for human embryo development in time lapse videos

Abstract Study question Can tracking cell division and predicting human embryo cleavage stages be automated in time-lapse videos (TLV) using AI object detection methods? Summary answer We developed software predicting blastomere count and tracking cell cleavages up until 4-5 stage. The software employs object detection technique called YOLOv5 to detect cells. What is known already Embryo morphology plays an important part in determining viability. Parameters such as number of cells present following fertilization, abnormal cell division (reverse/direct) and evaluating cleavage stages have correlation with pregnancy rates. However, continuous manual evaluation can be time-consuming, and automation will assist in embryo viability assessment. YOLOv5 has proven to accurately detect objects in videos. YOLOv5 uses mean average precision (mAP) as a metric to quantify the portions of frames in videos having the correct count of the objects. Study design, size, duration We have developed a software that uses YOLOv5 to detect cells present in frames of TLV, then marks each cell boundary with different colored circular overlays using OpenCV. We trained YOLOv5 to detect objects: cell, morula and blastocyst using 150 images of different cell-stages, morula, blastocyst. For object cell mAP was 0.65. Annotated location of objects in images and YOLOv5 predictions were reviewed by embryologists. We evaluated the software on TLV from 11 patients. Participants/materials, setting, methods After YOLOv5 detects cells in frames of TLV, our software computes cell count and assigns each cell a different color which is maintained until cell division into daughter cells. Later, daughter cells were also assigned different colors. If the frame has a preceding frame, software calculates detected cells' proximity with each cell in the preceding frame and copies color scheme provided proximity is within some threshold. The software provides TLV with colored overlays as output. Main results and the role of chance In starting frames of TLV with single cell, software accurately detected 1-cell (high precision=0.99, high recall=0.83, high F1-score=0.90). We observed some misclassification between 1-cell and morula. The reason could be that compacted morula looks like 1-cell. Best performance is observed for 2-cells (high precision=0.91, high recall=0.98, high F1-score=0.95). 4-cells were sometimes misclassified with 3 or 5-cells (high precision=0.88, low recall=0.59, high F1-score=0.71). One reason for the misclassification can be that overlapping between cells increases with number of cells. 3-cell and 5-cell are confused with other stages, still cleavage stage detection is better than random: 3-cell (average precision=0.43, high recall=0.83, average F1-score=0.49), 5-cell (average precision=0.44, average recall=0.40, average F1-score=0.40). For cell-stages>5, YOLOv5 detects less cells than actual count and software predicts cleavage later than actual by 9-10 frames on average. The proximity threshold used was 0.10 for cell-count<4 and 0.05 for count>4. In 5 TLV, overlay color for cells changes abruptly between frames, possibly because once YOLOv5 detected a stage, in consecutive frames less cell-number was recorded, and then again reported correct count. Sometimes, software selected the wrong parent for daughter cells (incorrect colored overlay). 2 TLV had direct and reverse cleavages and software could detect these two patterns. Limitations, reasons for caution Overall, our software can precisely detect cells, cell divisions and cleavage stages up to 4-cell stages. We hypothesize that training YOLOv5 on a bigger dataset and including several focal plane information will enable our software to detect overlapping cells and cleavage stages > =5. Wider implications of the findings Object detection proved to be pragmatic for ART and tracking cell division using our software will reduce time consumed in manual annotations, easier prediction of abnormal cleavages and more objective assessments. Qualitative evaluation by embryologists resulted in the overall verdict that this is useful and promising for further development. Trial registration number not applicable

Object Identification Model using Deep Reinforcement Machine Learning Concept for Image

This paper presents a model which gives the detailed process of object identification. We need to identify class and location of object in image for completing process of objet identification. Proposed model works on the principal of reinforcement learning which takes action on the basis of rewards and experiences. Normally methods in literature uses sliding window which moves in same direction but proposed algorithm provides a variable mask which moves 360 degree for identifying object using action history vector proposed with RL also not only this work focuses on localization like other work but also used class information with Softmax classification able to classify multiple object in single image with efficient time which is novel. Proposed mask acts as agent and focuses on proposed candidate reason this saves time and works in efficient manner for identification. Agent depends on transformation action and by applying top down reasoning it gives location of object. Classification is done using Softmax classification as we are having features of image by CNN. Reinforcement learning concept used for training of agent and Pascal voc dataset used for testing. Analysis of only 10 to 25 regions is sufficient with proposed work to identify first instance of object. Experiment and performance evaluation shows the efficiency of proposed work.

Learning to Localize Objects Using Limited Annotation, With Applications to Thoracic Diseases

Motivation: The localization of objects in images is a longstanding objective within the field of image processing. Most current techniques are based on machine learning approaches, which typically require careful annotation of training samples in the form of expensive bounding box labels. The need for such large-scale annotation has only been exacerbated by the widespread adoption of deep learning techniques within the image processing community: deep learning is notoriously data-hungry. Method: In this work, we attack this problem directly by providing a new method for learning to localize objects with limited annotation: most training images can simply be annotated with their whole image labels (and no bounding box), with only a small fraction marked with bounding boxes. The training is driven by a novel loss function, which is a continuous relaxation of a well-defined discrete formulation of weakly supervised learning. Care is taken to ensure that the loss is numerically well-posed. Additionally, we propose a neural network architecture which accounts for both patch dependence, through the use of Conditional Random Field layers, and shift-invariance, through the inclusion of anti-aliasing filters. Results: We demonstrate our method on the task of localizing thoracic diseases in chest X-ray images, achieving state-of-the-art performance on the ChestX-ray14 dataset. We further show that with a modicum of additional effort our technique can be extended from object localization to object detection, attaining high quality results on the Kaggle RSNA Pneumonia Detection Challenge. Conclusion: The technique presented in this paper has the potential to enable high accuracy localization in regimes in which annotated data is either scarce or expensive to acquire. Future work will focus on applying the ideas presented in this paper to the realm of semantic segmentation.

Weakly-Supervised Self-Training for Breast Cancer Localization.

The use of deep learning methods has dramatically increased the state-of-the-art performance in image object localization. However, commonly used supervised learning methods require large training datasets with pixel-level or bounding box annotations. Obtaining such fine-grained annotations is extremely costly, especially in the medical imaging domain. In this work, we propose a novel weakly supervised method for breast cancer localization. The essential advantage of our approach is that the model only requires image-level labels and uses a self-training strategy to refine the predicted localization in a step-wise manner. We evaluated our approach on a large, clinically relevant mammogram dataset. The results show that our model significantly improves performance compared to other methods trained similarly.

An Improved Deep Neural Network Model Based on Searching Space Limitation for Object Localization

Image object localization is the primary research direction in computer vision. A lots of algorithms had been proposed based on the ResNet structure and good results have been obtained on mainstream datasets. However, these methods still have the problem of too large ROI search space on weakly supervised data. This paper proposes a method that further limits the search space by directly extracting the high-level parameters of ResNet. The better search results and search efficiency are achieved and the candidate ROI search space is better reduced. The space information of the original images can be retained and training/testing time can be saved. The experimental results on the benchmark datasets show that this network structure has good object localization accuracy under weakly supervised data and the search efficiency is improved compared to other methods.

Crowdsourced object-labeling based on a game-based mobile application

Unparalleled growth in the sharing of media via networks has prompted a great deal of research into issues pertaining to image retrieval. The training and verification of image retrieval systems requires a large number of labelled images with ground truth; however, most researchers employ public datasets for their experiments, the results are restricted by the size and content of the dataset. In this study, we developed a system based on a mobile phone App for the collection of information pertaining to the location of objects in images. The proposed system is simple and easy to use. Experiments demonstrate the excellent performance of the proposed system with regard to accuracy and response time. This study demonstrates the feasibility of collecting image information using mobile phones.

Impact of bowtie filter and object position on the two-dimensional noise power spectrum of a clinical MDCT system.

Noise characteristics of clinical multidetector CT (MDCT) systems can be quantified by the noise power spectrum (NPS). Although the NPS of CT has been extensively studied in the past few decades, the joint impact of the bowtie filter and object position on the NPS has not been systematically investigated. This work studies the interplay of these two factors on the two dimensional (2D) local NPS of a clinical CT system that uses the filtered backprojection algorithm for image reconstruction. A generalized NPS model was developed to account for the impact of the bowtie filter and image object location in the scan field-of-view (SFOV). For a given bowtie filter, image object, and its location in the SFOV, the shape and rotational symmetries of the 2D local NPS were directly computed from the NPS model without going through the image reconstruction process. The obtained NPS was then compared with the measured NPSs from the reconstructed noise-only CT images in both numerical phantom simulation studies and experimental phantom studies using a clinical MDCT scanner. The shape and the associated symmetry of the 2D NPS were classified by borrowing the well-known atomic spectral symbols s, p, and d, which correspond to circular, dumbbell, and cloverleaf symmetries, respectively, of the wave function of electrons in an atom. Finally, simulated bar patterns were embedded into experimentally acquired noise backgrounds to demonstrate the impact of different NPS symmetries on the visual perception of the object. (1) For a central region in a centered cylindrical object, an s-wave symmetry was always present in the NPS, no matter whether the bowtie filter was present or not. In contrast, for a peripheral region in a centered object, the symmetry of its NPS was highly dependent on the bowtie filter, and both p-wave symmetry and d-wave symmetry were observed in the NPS. (2) For a centered region-ofinterest (ROI) in an off-centered object, the symmetry of its NPS was found to be different from that of a peripheral ROI in the centered object, even when the physical positions of the two ROIs relative to the isocenter were the same. (3) The potential clinical impact of the highly anisotropic NPS, caused by the interplay of the bowtie filter and position of the image object, was highlighted in images of specific bar patterns oriented at different angles. The visual perception of the bar patterns was found to be strongly dependent on their orientation. The NPS of CT depends strongly on the bowtie filter and object position. Even if the location of the ROI with respect to the isocenter is fixed, there can be different symmetries in the NPS, which depend on the object position and the size of the bowtie filter. For an isolated off-centered object, the NPS of its CT images cannot be represented by the NPS measured from a centered object.

Large Scale Learning for Food Image Classification

Since health care on foods is drawing people's attention recently, in this paper we propose a computer vision based food recognition system could be used to estimate food for diabetes patients. This study proposes a methodology for automatic food recognition, based on the Bag of Features (BoF) model. We present an approach to find out the group and location of objects in images. The system computes dense local features using scale invariant features. It performs very fast classification of each pixel in an image. For the design and valuation of the proposed system, a image dataset with nearly 5010 food images was created and organized into 11 classes. This system has achieved the accuracy of 78%.of objects in images. The system computes dense local features using scale invariant features. It performs very fast classification of each pixel in an image. For the design and valuation of the proposed system, a image dataset with nearly 5010 food images was created and organized into 11 classes. This system has achieved the accuracy of 78%. DOI: 10.17762/ijritcc2321-8169.150317

Localization of objects contours with different scales in images using hough transform

This paper considers the problem of localization of objects in images, when the scale of the sought for object is different from the scale of the sample object. We use the feature of the Hough transform to form the response in the shape of lines in the accumulator space; these lines correspond to the straight-line sections of the object’s contour. The paper contains the results of the formation of the responses in the accumulator space for some simple geometric figures. We introduce the technology of traffic signs localization, based on the considered approach.

Detection and localization of objects in images of 3D ground scenes

The problem of detecting and localizing given objects in 3D images of many-object ground scenes received by an onboard sensor of a pilotless vehicle is considered as a selection problem. The obtained algorithms employ a reference description of images of the observed scene, contours, and/or information about the angles of intensity jumps in pixels of the input image.