Narrow Results

Aiming at solving the problems of low recognition rate, low recognition efficiency and poor recognition effect in the current dance motion recognition methods that are affected by the surrounding environment, this study proposes a dance action recognition model based on the spatial frequency domain features of contour image. This study uses texture information to extract the dance action contour image, solve the feature vector of the contour image by the hypercomplex Fourier transform, and adopts the phase spectrum and energy spectrum transformation to smooth the contour image, so as to generate a saliency map, finally completes the extraction and preprocessing of dance action contour image. This paper distinguishes the high-frequency and low-frequency parts of dance action through the method of discrete cosine transform, calculates the number of pixels contained in the dance action images to be recognized, and extracts the spatial frequency domain features of contour image of dance action, builds the human posture model. This model realizes the dance action recognition by using the classifier to process the above-extracted feature vector and its label. The experimental result shows that the dance action recognition effect of this research model is good, and its recognition rate is high in different dance action types, and can effectively meet the needs of dance action recognition.

Breast cancer (BrC) is one of the most common causes of death among women worldwide. Images of the breast (mammography or ultrasound) may show an anomaly that represents early indicators of BrC. However, accurate breast image interpretation necessitates labor-intensive procedures and highly skilled medical professionals. As a second opinion for the physician, deep learning (DL) tools can be useful for the diagnosis and classification of malignant and benign lesions. However, due to the lack of interpretability of DL algorithms, it is not easy to understand by experts as to how to predict a label. In this work, we proposed multitask U-Net Saliency estimation and DL model-based breast lesion segmentation and classification using ultrasound images. A new contrast enhancement technique is proposed to improve the quality of original images. After that, a new technique was proposed called UNET-Saliency map for the segmentation of breast lesions. Simultaneously, a MobileNetV2 deep model is fine-tuned with additional residual blocks and trained from scratch using original and enhanced images. The purpose of additional blocks is to reduce the number of parameters and better learning of ultrasound images. Training is performed from scratch and extracted features from the deeper layers of both models. In the later step, a new cross-entropy controlled sine-cosine algorithm is developed and selected best features. The main purpose of this step is the reduction of irrelevant features for the classification phase. The selected features are fused in the next step by employing a serial-based Manhattan Distance (SbMD) approach and classified the resultant vector using machine learning classifiers. The results indicate that a wide neural network (W-NN) obtained the highest accuracy of 98.9% and sensitivity rate of 98.70% on the selected breast ultrasound image dataset. The comparison of the proposed method accuracy is conducted with state-of-the-art (SoArt) techniques which show the improved performance.

Saliency region detection methods have become one of the hotspots in the field of image processing as an important method to improve the real-time and accurate analysis of massive data. Integrating more effective prior knowledge is a viable direction for improving the performance of saliency region detection methods. Most of the methods based on background prior and boundary connectivity prior assume the boundary area of the image as the background, by restraining the background to highlight the salient area. When the boundary area of the image does not describe the background well (such as a large difference in border area features), if the entire frame of the image is put together to compute the background feature, the calculation of the background feature will be inaccurate. In view of the above shortcomings, this paper proposed a saliency region detection method based on background and spatial position. This method carried on the image boundary super pixel clustering, determined the background feature according to the clustering center, and used the difference between the super pixel on the image and the background super pixel, and its spatial position to calculate the salient of the super pixels. This approach used MATLAB to program and experiment. The method was compared with a series of the state-of-the-art methods. The AUC of proposed algorithm reaches 0.839, and the MAE is 0.220, showing the effectiveness of the proposed algorithm.

In this paper, we propose a prior fusion and feature transformation-based principal component analysis (PCA) method for saliency detection. It relies on the inner statistics of the patches in the image for identifying unique patterns, and all the processes are done only once. First, three low-level priors are incorporated and act as guidance cues in the model; second, to ensure the validity of PCA distinctness model, a linear transform for the feature space is designed and needs to be trained; furthermore, an extended optimization framework is utilized to generate a smoothed saliency map based on the consistency of the adjacent patches. We compare three versions of our model with seven previous methods and test them on several benchmark datasets. Different kinds of strategies are adopted to evaluate the performance and the results demonstrate that our model achieves the state-of-the-art performance.

Saliency detection refers to the segmentation of all visually conspicuous objects from various backgrounds. The purpose is to produce an object-mask that overlaps the salient regions annotated by human vision. In this paper, we propose an efficient bottom-up saliency detection model based on wavelet generalized lifting. It requires no kernels with implicit assumptions and prior knowledge. Multiscale wavelet analysis is performed on broadly tuned color feature channels to include a wide range of spatial-frequency information. A nonlinear wavelet filter bank is designed to emphasize the wavelet coefficients, and then a saliency map is obtained through linear combination of the enhanced wavelet coefficients. This full-resolution saliency map uniformly highlights multiple salient objects of different sizes and shapes. An object-mask is constructed by the adaptive thresholding scheme on the saliency maps. Experimental results show that the proposed model outperforms the existing state-of-the-art competitors on two benchmark datasets.

Computational modeling of the human visual system is of current interest to developers of artificial vision systems, primarily because a biologically-inspired model can offer solutions to otherwise intractable image understanding problems. The purpose of this study is to present a biologically-inspired model of selective perception that augments a stimulus-driven approach with a high-level algorithm that takes into account particularly informative regions in the scene. The representation is compact and given in the form of a topographic map of relative perceptual conspicuity values. Other recent attempts at compact scene representation consider only low-level information that codes salient features such as color, edge, and luminance values. The previous attempts do not correlate well with subjects' fixation locations during viewing of complex images or natural scenes. This study uses high-level information in the form of figure/ground segmentation, potential object detection, and task-specific location bias. The results correlate well with the fixation densities of human viewers of natural scenes, and can be used as a preprocessing module for image understanding or intelligent surveillance applications.

Video object segmentation of real-world scenes is a challenging task due to the dynamic environment change such as uneven illumination, object articulation, as well as camera motion. In this paper, we proposed a method to combine semantic regions by trained deep convolutional neural networks with saliency map and motion cues to segment foreground objects in video sequences. Firstly, the parameters of the deep convolutional networks are trained using PASCAL VOC dataset with human annotations. The training process consists of forward inference and background learning stages. The learning process employs the standard stochastic gradient descending algorithm. The number of epochs during training process is fixed at 7. Secondly, the trained convolutional networks are used to predict the semantic labels of a real-world video sequence at per-frame level. The inferenced semantic region is combined with the saliency map through Markov random field to derive foreground objects at each frame of the video. To access the segmentation performance of the proposed algorithm, we test the proposed algorithm using a video sequence from FBMS motion segmentation benchmarks and compare the segmentation accuracy with state-of-the-art video object segmentation algorithms.

Objective image quality assessment (IQA) metrics have been widely applied to imaging systems to preserve and enhance the perceptual quality of images being processed and transmitted. In this paper, we present a novel IQA metric based on biologically inspired feature model (BIFM) and structural similarity index (SSIM). The SSIM index map is first generated through the well-known IQA metric SSIM between the reference image and the distorted image. Then, saliency map of the distorted image is extracted via BIF to define the most salient image locations. Finally, according to the saliency map, a feature weighting model is employed to define the different weights for the different samples in the SSIM index map. Experimental results confirm that the proposed IQA metric improves the performance over PSNR and SSIM under various distortion types in terms of different evaluation criteria.

Content-based image retrieval (CBIR) makes use of low-level image features, such as color, texture and shape, to index images with minimal human interaction. Considering the gap between low-level image features and the high-level semantic concepts in the CBIR, we proposed an image retrieval system for brain magnetic resonance images based on saliency map. The saliency map of an image contains important image regions which are visually more conspicuous by virtue of their contrast with respect to surrounding regions. First, the proposed approach exploits the ant colony optimization (ACO) technique to measure the image's saliency through ants' movements on the image. The textural features are then calculated from the saliency map of the images. The image indexing is done with an adaptive neuro-fuzzy inference system (ANFIS), which can categorize the magnetic resonance images as normal or tumoral. In online image retrieval, a query image is introduced to the system and the system will return the relevant images. The experimental result shows the accuracy of 98.67% for the image retrieval in our proposed system and improves the retrieval efficiency in compare with the classical CBIR systems.

Content-based image retrieval (CBIR) has turned into an important research field with the advancement in multimedia and imaging technology. The term CBIR has been widely used to describe the process of retrieving desired images from a large collection on the basis of features such as color, texture and shape that can be automatically extracted from the images themselves. Considering the gap between low-level image features and the high-level semantic concepts in the CBIR, we proposed an image retrieval system for brain magnetic resonance images based on saliency map. First, the proposed approach exploits the ant colony optimization (ACO) technique to measure the image’s saliency through ants’ movements on the image. The textural features are then calculated from the saliency map of the images. The image retrieval of the proposed CBIR system is based on textural features and the fuzzy approach using Adaptive neuro-fuzzy inference system (ANFIS). Regarding the various categories of images in a database, we define some membership functions in the ANFIS output in order to determine the membership values of the images related to the existing categories. In online image retrieval, a query image is introduced to the system and the relevant images can be retrieved based on query membership values into different classes including normal or tumoral. The experimental results indicate that the proposed method is reliable and has high image retrieval efficiency compared with the previous works.

In the field of image saliency detection, previous approaches are mostly built on the low level priors like color and space contrast, boundary or connectivity prior, which are not sufficient to differentiate real salient regions from other independent or high-contrast parts. In this paper, we propose a novel approach which utilizes both high level and low level priors or cues to generate a salient map. Specifically, we obtain a foreground likelihood map via low rank matrix recovery, which incorporates traditional low-level features with higher-level guidance. Then, we compute a background likelihood map via principal component analysis and k-means clustering, which utilize two low level priors: distribution and boundary priors. Finally, the salient values of one image are calculated based on the two likelihood maps via a modified optimization framework. Extensive experiments on numerous publicly available datasets demonstrate that our proposed algorithm outperforms state-of-the-art methods.

In urban search and rescue mobile robotics, one of the most significant problems is to identify and localize trapped victims in collapsed or dangerous buildings. In order to optimize search operations, selective visual attention plays a crucial role, as it, focuses on the visual stimuli that are most relevant when identifying victims, inhibiting other stimuli in the scene. This reduces the amount of resources required for subsequent visual processing, such as recognition, and typically results in more effective and quicker search strategies. In this work, we concentrate on a purely bottom-up computational model. It functions by differentiating high-contrast areas in the image using several features including intensity, colour, and visual motion. Results for artificial and real-world images are presented, showing that the implemented system has an acceptable performance in identifying moving human body parts.

In recent years, the target detection based on Human Visual System (HVS) has been extensively studied by scholars at home and abroad, especially for the visible target detection. Generally, these algorithms involve three processes, the extraction of visual feature according to the targets' characteristics, the generation of saliency map to segment the targets and the simulation of eye movement to track the targets. Considering the small infrared targets have less information than visible targets, such as color and shape, but they have saliency characteristic of contrast, in this paper, an algorithm of small infrared target detection based on visual contrast mechanism has been proposed. Firstly, according to the targets' local contrast feature, the saliency map of the image is generated. By calculating the saliency map, it can enhance the targets and suppress the background. Then the targets can be detected after the saliency map is segmented. Considering it is difficult to detect the targets when the image's signal-to-noise ratio (SNR) is very low even the targets are submerged in the background, the algorithm simulates eye movement to track the targets. The position of the suspicious target is predicted by the algorithm of Proportional-Integral-Derivative (PID). By enhancing the suspicious target's local region using the Retinex theory and segmenting the local region, the targets can be re-detected. By comparing the proposed algorithm with other methods, the experiments show that the proposed method works well, not only it can be applied to the situation the target is submerged in the background, but it can also be used in different complex background.