Feature Enhancement in Visually Impaired Images

Synthetic Iris Images: A Comparative Analysis between Cartesian and Polar Representation

In recent years, the advancement of generative techniques, particularly generative adversarial networks (GANs), has opened new possibilities for generating synthetic biometric data from different modalities, including-among others-images of irises, fingerprints, or faces in different representations. This study presents the process of generating synthetic images of human irises, using the recent StyleGAN3 model. The novelty presented in this work consists in producing generated content in both Cartesian and polar coordinate representations, typically used in iris recognition pipelines, such as the foundational work proposed by John Daugman, but hitherto not used in generative AI experiments. The main objective of this study was to conduct a qualitative analysis of the synthetic samples and evaluate the iris texture density and suitability for meaningful feature extraction. During this study, a total of 1327 unique irises were generated, and experimental results carried out using the well-known OSIRIS open-source iris recognition software and the equivalent software, wordlcoin-openiris, newly published at the end of 2023 to prove that (1) no "identity leak" from the training set was observed, and (2) the generated irises had enough unique textural information to be successfully differentiated between both themselves and between them and real, authentic iris samples. The results of our research demonstrate the promising potential of synthetic iris data generation as a valuable tool for augmenting training datasets and improving the overall performance of iris recognition systems. By exploring the synthetic data in both Cartesian and polar representations, we aim to understand the benefits and limitations of each approach and their implications for biometric applications. The findings suggest that synthetic iris data can significantly contribute to the advancement of iris recognition technology, enhancing its accuracy and robustness in real-world scenarios by greatly augmenting the possibilities to gather large and diversified training datasets.

High-precision visual localization based on an improved shape-based matching algorithm

To deal with the requirement of high-precision localization of large-size workpieces in an industrial environment, an improved shape-based matching algorithm is proposed based on the phase stretching transformation and the iterative closest point algorithms. Basler industrial cameras are used to collect images of large-size workpieces, such as glass. The experimental results show that the average localization error is 0.05±0.013mm, which can meet the requirements of practical applications. This algorithm can effectively and accurately achieve high-precision localization of different positions of multi-directionally transformed objects in industrial environments.

An Image Enhancement Algorithm Based on Fractional-Order Phase Stretch Transform and Relative Total Variation

The main purpose of image enhancement technology is to improve the quality of the image to better assist those activities of daily life that are widely dependent on it like healthcare, industries, education, and surveillance. Due to the influence of complex environments, there are risks of insufficient detail and low contrast in some images. Existing enhancement algorithms are prone to overexposure and improper detail processing. This paper attempts to improve the treatment effect of Phase Stretch Transform (PST) on the information of low and medium frequencies. For this purpose, an image enhancement algorithm on the basis of fractional-order PST and relative total variation (FOPSTRTV) is developed to address the task. In this algorithm, the noise in the original image is removed by low-pass filtering, the edges of images are extracted by fractional-order PST, and then the images are fused with extracted edges through RTV. Finally, extensive experiments were used to verify the effect of the proposed algorithm with different datasets.

Image segmentation of activated sludge phase contrast images using phase stretch transform

Activated sludge (AS) is a biological treatment process that is employed in wastewater treatment plants. Filamentous bacteria in AS plays an important role in the settling ability of the sludge. Proper settling of the sludge is essential for normal functionality of the wastewater plants, where filamentous bulking is always a persistent problem preventing sludge from settling. The performance of AS plants is conventionally monitored by physico-chemical procedures. An alternative way of monitoring the AS in wastewater treatment process is to use image processing and analysis. Good performance of the image segmentation algorithms is important to quantify flocs and filaments in AS. In this article, an algorithm is proposed to perform segmentation of filaments in the phase contrast images using phase stretch transform. Different values of strength (S) and warp (W) are tested to obtain optimum segmentation results and decrease the halo and shade-off artefacts encountered in phase contrast microscopy. The performance of the algorithm is assessed using DICE coefficient, accuracy, false positive rate (FPR), false negative rate (FNR) and Rand index (RI). Sixty-one gold approximations of ground truth images were manually prepared to assess the segmentation results. Thirty-two of them were acquired at 10× magnification and 29 of them were acquired at 20× magnification. The proposed algorithm exhibits better segmentation performance with an average DICE coefficient equal to 52.25%, accuracy 99.74%, FNR 41.8% and FPR 0.14% and RI 99.49%, based on 61 images.