International Journal of Computer and Applications

An imaging apparatus and method are provided configured for discriminating between sky and non-sky regions The apparatus comprises a UV-passing image forming element and a UV-sensing image capturing sensor for capturing an image projected from the UV- passing image forming element The image comprising at least one of sky and non-sky image regions and the image captured by the sensor is provided to an image processing unit for performing discrimination between sky and non-sky regions of the captured image using primarily UV wavelengths

Image capture apparatus and method

Inspired by ant navigation, we explore a method for sky segmentation using ultraviolet (UV) light. A standard camera is adapted to allow collection of outdoor images containing light in the visible range, in UV only and in green only. Automatic segmentation of the sky region using UV only is significantly more accurate and far more consistent than visible wavelengths over a wide range of locations, times and weather conditions, and can be accomplished with a very low complexity algorithm. We apply this method to obtain compact binary (sky vs non-sky) images from panoramic UV images taken along a 2km route in an urban environment. Using either sequence SLAM or a visual compass on these images produces reliable localisation and orientation on a subsequent traversal of the route under different weather conditions.

/pdf/sky-segmentation-with-ultraviolet-images-can-be-used-for-lqhxsmxw94.pdf

Sky segmentation with ultraviolet images can be used for navigation

In this paper, a novel watershed approach based on seed region growing and image entropy is presented which could improve the medical image segmentation. The proposed algorithm enables the prior information of seed region growing and image entropy in its calculation. The algorithm starts by partitioning the image into several levels of intensity using watershed multi-degree immersion process. The levels of intensity are the input to a computationally efficient seed region segmentation process which produces the initial partitioning of the image regions. These regions are fed to entropy procedure to carry out a suitable merging which produces the final segmentation. The latter process uses a region-based similarity representation of the image regions to decide whether regions can be merged. The region is isolated from the level and the residual pixels are uploaded to the next level and so on, we recall this process as multi-level process and the watershed is called multi-level watershed. The proposed algor...

A watershed approach for improving medical image segmentation.

Low cost implementation of an accurate face recognition system is useful in many applications. A face recognition algorithm based on PCA and wavelets is proposed. Effectiveness of using wavelets before PCA is shown by performing comparison of different simulation results. Complete hardware implementation of the recognition phase is then proposed. The main contribution of this work is the low execution time and minimal internal FPGA consumed resources. The proposed architecture includes three building blocks of the recognition phase. First, a wavelet transform of the image is generated using the convolution version, then, the projection in the eigenspace and the distance. It is shown that the entire face recognition system can be implemented on FPGA virtex II PRO. This implementation shows acceptable hardware resources occupation and a maximum frequency of 46.794MHz for a 60×60 image.

Implementation of face recognition system in virtex II Pro platform

algorithm as was introduced by Vincent and Soille is a segmentation algorithm based on the inundation process of the image gradient which is observed as a relief. It aims at finding the peaks in the image gradient called watersheds and identifying them as the image contours. Due to its flexibility and rapidity, this algorithm is used in several applications. However, its main drawback is the over segmentation .In this paper, we improve this technique by introducing a histogram driven methodology. The developed architecture is applied on an empirical basis for research on image segmentation and boundary detection in order to be compared with other segmentation algorithms. The simulation results show that the performance of our algorithm is superior to the other segmentation techniques. Finally, the whole design is implemented on a Virtex 5 platform based on a co- design methodology leading to 147 MHz frequency and 76% of hardware resource occupation for an image of the size of 128*128.

/pdf/improving-watershed-algorithm-with-a-histogram-driven-1rjnxmtb73.pdf

Improving Watershed Algorithm with a Histogram Driven Methodology and Implementation of the System on a Virtex 5 Platform

Watershed transformation is a powerful technique that can be efficiently used for image segmentation. In this paper, we implement a watershed based segmentation algorithm on a Virtex II pro platform. The main contribution of this work is the low execution time and minimal internal FPGA consumed resources. 'The proposed architecture includes two main blocs. First, a gradient of the image is generated using morphological operators (dilation-erosion). Then, Watershed is applied to the resultant image based on immersion principle. The proposed implementation was optimized with respect to hardware resources occupation and speed, based on a Codesign methodology. Our approach makes use of the potential software in Virtex II PRO platform enabled by the embedded processor power PC. Firstly, in a high design level, the whole design was done in PowerPC. Then, the optimal design is carried out by analyzing the timing of the different portions of the algorithm and implementing the time extensive parts as hardware This strategy leads to acceptable hardware resources occupation and a maximum frequency performance of approximately 100 MHz. As illustration, we apply our design to the segmentation of Cameraman image.

Implementation of watershed based image segmentation algorithm in virtex II pro platform

Watershed transform, as was described by Vincent and Soille, is a segmentation algorithm based on a flooding process of the gradient image, which is observed as a topographic surface. Watershed transform aims at finding the peaks in this surface and identifying them as image contours. This algorithm is used in many applications for its flexibility. However, its main limitation is the over segmentation. In this paper, we will try to overcome this limitation. The enhancement phase is based on filtering the original image by an anisotropic diffusion filter and then quantizing the gradient image. For evaluation, the improved algorithm is applied on an empirical basis for image segmentation research in order to be compared with other segmentation algorithms. Simulation results prove the effectiveness of our algorithm. Watershed algorithm is used in medical field in order to extract areas of interest in an image representing a section bone. Finally, the implementation phase in a Virtex 5 is based on the co design methodology. The synthesis results show that the performance of the developed design is 130 MHz and the hardware occupation is about 78% for an image of size of 256*256.

/pdf/implementation-of-an-improved-watershed-algorithm-in-a-2jlswwqhek.pdf

Implementation of an Improved Watershed Algorithm in a Virtex 5 Platform

Face Recognition (FR) has been widely used in the tracking and identification of individuals. However, because face images vary depending on expressions, ages, individual locations, and lighting conditions, the facial photographs of the same sample may appear to be distinct, making face recognition more difficult. Deep learning (DL) is now a suitable solution for face recognition and computer vision. In this study, features and traits were extracted from images of a large data set (called FERET) consisting of 14,126 images that were divided into 80% for training data and 20% for testing data using a Convolutional Neural Network (CNN). The CNN is first pre-trained using supplementary data for the purpose of obtaining updated weights, and then trained with the target dataset in order to uncover more hidden facial characteristics. Three different deep learning models are implemented: AlexNet, Resnet18, and DenseNet-161. The performance of these models is compared experimentally in terms of their classification accuracy. The obtained results showed that the DenseNet-161 has the highest accuracy of 98.6%, while the accuracies of the Resnet18 and AlexNet are 96.3% and 93.3%, respectively.

Face Recognition in Multiple Variations Using Deep Learning and Convolutional Neural Networks

Breast cancer is one of the most important diseases that lead to death, according to the reports of the World Health Organization. Reports also indicated that breast cancer affects women more than men. Late or wrong diagnosis leads to a deterioration in the patient's condition and may lead to death. Therefore, automated multiple classifications using machine learning of breast cancer from histopathological images plays a major role in the early diagnosis of the disease. Multiple classifications of breast cancer is the identification of tumors such as Carcinoma, Fibroadenoma, Cysts, etc. However, this type of classification faces many challenges such as extracting subtle differences between the binary classes of tumors and whether they are benign or malignant, strong coherence of tumor cells and widespread color heterogeneity. Therefore, we propose in this paper the use of a sophisticated model that works on semantic segmentation and extracting distinctive patterns and classifying them using a deep neural network. The structured deep learning model performed impressively (average accuracy of 98.7%) on a large-scale data set, demonstrating the power of our method in providing an effective tool for classifying multiple breast cancers in clinical settings.

Nadia Smaoui Zghal

Papers

Improving Watershed Algorithm with a Histogram Driven Methodology and Implementation of the System on a Virtex 5 Platform

Implementation of watershed based image segmentation algorithm in virtex II pro platform

Implementation of an Improved Watershed Algorithm in a Virtex 5 Platform

Face Recognition in Multiple Variations Using Deep Learning and Convolutional Neural Networks

Multiclassification Model of Histopathological Breast Cancer Based on Deep Neural Network