Haniza Yazid
Thumbnail Image
Preferred name
Haniza Yazid
Official Name
Haniza, Yazid
Alternative Name
Yazid, H.
Yazid, Haniza
Main Affiliation
Scopus Author ID
22137274300
Researcher ID
D-3830-2015

Research Output

Filters

3 2 2
3
Reset filters

Settings
Now showing 1 - 3 of 3
Thumbnail Image
Publication

Object detection using image processing techniques: coconut as a case study

2007 , Haniza Yazid

The use of computers to analyze images has many potential but, the variability of the objects makes it a challenging task. In this thesis, the main idea is to detect an object (coconut) from an image. Several techniques have been utilized namely, the separable filter, Circular Hough Transform (CHT), chord intersection and moment invariant. Before applying these techniques, the preprocessing and image segmentation steps need to be performed in priori. Histogram equalization is utilized in preprocessing step meanwhile edge detection and morphological filtering have been employed in image segmentation step. Single object has been experimented to evaluate the two (2) techniques, CHT and the chord intersection. Based on the results obtained from single object detection, the CRT achieves higher percentage, 87.5% than chord intersection technique, 85%. For multiple objects detection, the CHT technique has been used and the highest detection for the first object is 87.5% followed by 92.5% for the second object, 77.5% for the third object and the last object is 67.5%. The moment invariant technique has been used to extract the shape of the object and detect its presence. From 50 images that have been experimented, 90% show positive result. This research can be adopted for climbing robotic system that can automatically pluck the coconut from a tree. Using image processing techniques, the gripping process will be easier and convenient than manual plucking.

View
No Thumbnail Available
Publication

Analysis on Single-Image Super-Resolution (SISR) Using Dictionary Learning and Sparse Representation Algorithm

2022-01-01 , Ng S.M. , Haniza Yazid , Mustafa N.

Image Super-Resolution (SR) is a technique in order to produce High-Resolution (HR) image from the corresponding Low-Resolution (LR) image by removing the degradation caused by imaging process of LR camera. In this work, a Single-Image Super-Resolution (SISR) image reconstruction scheme based on dictionary learning process with sparse representation method is proposed. As a result, the image quality of the obtained HR image decreased significantly with increasing of the upscale factor. Then, the analysis showed that the HR image obtained by applying the proposed work was able to produce a better performance in terms of Root Mean Square Error (RMSE), Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Matric (SSIM) values as compared to the bicubic interpolation operation. Therefore, the work done in this paper is able to solve the LR problem in images by proposing a SISR image reconstruction scheme based on dictionary learning process with sparse representation algorithm. Lastly, this work can be improved by testing on different types of images such as biometric images.

View
No Thumbnail Available
Publication

Performance analysis on dictionary learning and sparse representation algorithms

2022-05-01 , Ng S.M. , Haniza Yazid , Mustafa N.

Theoretically, the Super-Resolution (SR) reconstruction scheme is a method which is performed by many applications nowadays for the purpose of generating a High-Resolution (HR) image using the input Low-Resolution (LR) images by filling in the missing high frequency information. In addition, the SR reconstruction implemented based on the theory of sparse representation techniques is known as an effective way to produce HR images using images patches generated from the LR images. In order to improve the quality of denoised images produced by using the sparse representation techniques, a scheme called dictionary learning algorithms could be considered. Thus, the objective of this paper is to provide a performance comparison on the effectiveness of applying the dictionary learning steps with sparse representation algorithms in producing a better denoised image. In this case, the average Peak Signal-to-Noise ratio (PSNR) and Structural Similarity Index Metric (SSIM) values of the denoised image obtained by using Algorithms 1, 2, and 3 which combined the use of dictionary learning and sparse representation algorithms were compared with the values obtained from images produced by applying only sparse regularisation methods. As a conclusion, the denoised images produced by Algorithm 1 in this paper had the greatest average PSNR and SSIM values. Hence, the algorithm with the implementation of the dictionary learning process with sparse representation methods is able to achieve a better result in enhancing the low-resolution images.

View