With the rapid development of smart devices, people's lives have become easier, especially for visually disabled or special-needs people. The new achievements in the fields of machine learning and deep learning let people identify and recognise the surrounding environment. In this study, the efficiency and high performance of deep learning architecture are used to build an image classification system in both indoor and outdoor environments. The proposed methodology starts with collecting two datasets (indoor and outdoor) from different separate datasets. In the second step, the collected dataset is split into training, validation, and test sets. The pre-trained GoogleNet and MobileNet-V2 models are trained using the indoor and outdoor se

     The searching process using a binary codebook of combined Block Truncation Coding (BTC) method and Vector Quantization (VQ), i.e. a full codebook search for each input image vector to find the best matched code word in the codebook, requires a long time.   Therefore, in this paper, after designing a small binary codebook, we adopted a new method by rotating each binary code word in this codebook into 900 to 2700 step 900 directions. Then, we systematized each code word depending on its angle  to involve four types of binary code books (i.e. Pour when , Flat when  , Vertical when, or Zigzag). The proposed scheme was used for decreasing the time of the coding procedure, with very small distortion per block, by designing s

The effect of using three different interpolation methods (nearest neighbour, linear and non-linear) on a 3D sinogram to restore the missing data due to using angular difference greater than 1° (considered as optimum 3D sinogram) is presented. Two reconstruction methods are adopted in this study, the back-projection method and Fourier slice theorem method, from the results the second reconstruction proven to be a promising reconstruction with the linear interpolation method when the angular difference is less than 20°.

FG Mohammed, HM Al-Dabbas, Iraqi journal of science, 2018 - Cited by 6

Semantic segmentation is an exciting research topic in medical image analysis because it aims to detect objects in medical images. In recent years, approaches based on deep learning have shown a more reliable performance than traditional approaches in medical image segmentation. The U-Net network is one of the most successful end-to-end convolutional neural networks (CNNs) presented for medical image segmentation. This paper proposes a multiscale Residual Dilated convolution neural network (MSRD-UNet) based on U-Net. MSRD-UNet replaced the traditional convolution block with a novel deeper block that fuses multi-layer features using dilated and residual convolution. In addition, the squeeze and execution attention mechanism (SE) and the s

In this paper, CdS/Si hetrojunction solar cell has been made by
Chemical Bath Deposition (CBD) of CdS thin film on to
monocrystalline silicon substrate. XRD measurements approved that
CdS film is changing the structure of CdS films from mixed
hexagonal and cubic phase to the hexagonal phase with [101]
predominant orientation. I-V characterization of the hetrojunction
shows good rectification, with high spectral responsivity of 0.41
A/W, quantum efficiency 90%,and specific detectivity 2.9*1014
cmHz1/2W -1 .

Data hiding is the process of encoding extra information in an image by making small modification to its pixels. To be practical, the hidden data must be perceptually invisible yet robust to common signal processing operations. This paper introduces a scheme for hiding a signature image that could be as much as 25% of the host image data and hence could be used both in digital watermarking as well as image/data hiding. The proposed algorithm uses orthogonal discrete wavelet transforms with two zero moments and with improved time localization called discrete slantlet transform for both host and signature image. A scaling factor ? in frequency domain control the quality of the watermarked images. Experimental results of signature image

This paper provides an attempt for modeling rate of penetration (ROP) for an Iraqi oil field with aid of mud logging data. Data of Umm Radhuma formation was selected for this modeling. These data include weight on bit, rotary speed, flow rate and mud density. A statistical approach was applied on these data for improving rate of penetration modeling. As result, an empirical linear ROP model has been developed with good fitness when compared with actual data. Also, a nonlinear regression analysis of different forms was attempted, and the results showed that the power model has good predicting capability with respect to other forms.