A novel method of crowd estimation is proposed in this paper: Firstly, surveillance image is divided into bit planes by ostu algorithm, the pixel ratio of foreground to background and complexity of bit planes are take...
详细信息
ISBN:
(纸本)9781424446902
A novel method of crowd estimation is proposed in this paper: Firstly, surveillance image is divided into bit planes by ostu algorithm, the pixel ratio of foreground to background and complexity of bit planes are taken as feature vectors of crowd estimation. The degree of crowd density of the scene is classified into several grades, BP neural network is used for training and then the classification model is constructed, through which the estimation of crowd density can be obtained. Experiments were taken based on video of two real scenes, the result show that this proposed approach is able to judge the levels of congestion with accuracy higher than 85%.
Fractal dimension is an important quantitative characteristic of a image, which can be widely used in image analysis. Differential box-counting method which is one of many calculation methods of a fractal dimension ha...
详细信息
ISBN:
(纸本)9781510601673
Fractal dimension is an important quantitative characteristic of a image, which can be widely used in image analysis. Differential box-counting method which is one of many calculation methods of a fractal dimension has been frequently used due to its simple calculation. In differential box-counting method, a window size M is limited in the integer power of 2. It leads to inaccurate calculation results of a fractal dimension. Aiming at solving the issues, in this paper, an improved algorithm is discussed that the window size M has been improved to be able to accommodate non-integer power of 2, and making the calculated fractal dimension error smaller. In order to verify superiority of the improved algorithm, the values of fractal dimension are regarded as parameters, and are applied for image segmentation combined with ostu algorithm. Both traditional and improved differential box-counting methods are respectively used to estimate fractal dimensions and do threshold segmentation for a thread image. The experimental results show that image segmentation details by improved differential box-counting method are more obvious than that by traditional differential box-counting method, with less impurities, clearer target outline and better segmentation effect.
Ultrasound Computed Tomography (USCT) of soft biological tissues today provides images with a high-level of resolution. The signal acquisition system using multichannel and/or multifrequency arrays performs in circula...
详细信息
ISBN:
(数字)9781510625587
ISBN:
(纸本)9781510625587
Ultrasound Computed Tomography (USCT) of soft biological tissues today provides images with a high-level of resolution. The signal acquisition system using multichannel and/or multifrequency arrays performs in circular mode and the main (linear) inversion algorithms are based on compression wave propagation modeling. The main limits of these methods for bone imaging are due to the large impedance contrast between tissues, and to propagative phenomena generated through periosteal interfaces (mode conversion, attenuation). The linear inversion methods fail to provide high-level resolution images. Despite their performance and robustness, the non-linear methods are still today unsuitable for clinical applications because of the high computation time required. However, in the special case of children bone imaging, acquisition steps must be as fast as possible, with short-time exposure and low-intensity sonication. In this context, we have developed a fast-acquisition setup (1 sec.) based on a cylindrical-focusing ring antenna, and a protocol (<5 sec.) using classical Born approximation and spatial Fourier transform. Unfortunately, the result today is a poor contrast-to-noise ratio (CNR) image. Previous work done to improve CNR used signal and image processing. This work focuses on this last point, and an automatic edge detection procedure, using Haar wavelet 2D-decompositon, combining k-means and ostu algorithms. Results will be presented on ex vivo real bone samples and on geometrical mimicking bone phantom (Sawbones (TM)). An example of bone defect imaging will be presented and discussed.
Strawberry picking by machines confronts a complex environment where a targe may be sheltered by leaves or overlap with each other. Also, it is a challenge for machines to recognize mature straw berries among those in...
详细信息
ISBN:
(纸本)9781538672556
Strawberry picking by machines confronts a complex environment where a targe may be sheltered by leaves or overlap with each other. Also, it is a challenge for machines to recognize mature straw berries among those in different maturity. This work presents a fast recognition method for elevated mature strawberries by the approach of deep learning. It uses an ostu algorithm to separate targets from background and then the resulted effective image areas designated by the minimum external rectangular marking method are used to train CaffeNet for automatic target recognition. For comparison, we also design a SVM classifer that uses HOG gradient direction feature and H component of the color feature of the mature strawberries. The experimental results show that the average recognition rate of mature strawberries by CaffeNet can reach 95%, higher than that by SVM by 11%.
As the mainstream products of flat panel display, LED screen is becoming more and more popular, which has been widely used in the display system of large gymnasium. The detection of LED display system in large stadium...
详细信息
As the mainstream products of flat panel display, LED screen is becoming more and more popular, which has been widely used in the display system of large gymnasium. The detection of LED display system in large stadium is analyzed. Based on the large LED display in Jingzhou stadium, the qualification of LED display system after installation is tested. A scanning line seed filling algorithm is proposed and used to collect the gray value in the connected domain of the lamp and obtain the average gray value of the light point, so that the LED display system of the large gymnasium is detected. The experimental results show that a reasonable LED display and adjustment scheme can ensure the installation of LED display system in large gymnasium in line with the requirements of the national standard.
A new kind of periodic randomized adaptive particle swarm algorithm,which can get fast convergence rate and effectively break away from the local extremum,can manifest more stronger optimization *** segmentation thres...
详细信息
A new kind of periodic randomized adaptive particle swarm algorithm,which can get fast convergence rate and effectively break away from the local extremum,can manifest more stronger optimization *** segmentation threshold should be fixed,using new particle swarm optimization algorithm could obtain optimal *** with the Grayscale median algorithm and ostu algorithm based on lena picture,the new method can get better results among all image segmentation results.
In non-contract anthropometric measurements, the accurate location of body feature points is the basis of body dimension data. The traditional corner based feature point location method may involve errors due to the b...
详细信息
In non-contract anthropometric measurements, the accurate location of body feature points is the basis of body dimension data. The traditional corner based feature point location method may involve errors due to the body shape variations. In this paper, we propose a novel body feature points location method. First, the human body contour is extracted with Otsu algorithm and Canny edge detection algorithm. Then, human height is obtained by horizontal scanning algorithm. Furthermore, we propose the shoulder location algorithm based on shoulder slope and curvature to locate the shoulder points. Finally, the rest body feature points are determined with three factors: the proportional relationship between body parts and height, corners and symmetry of human body contour. The experimental results show that the proposed method can accurately locate the body feature points and solve the issue of the loss of feature points of human waist and chest.
暂无评论