Orthogonal Rotation-Invariant Moments for Digital Image Processing

Orthogonal rotation invariant moments (ORIMs) exhibit attractive characteristics such as rotation invariance (translation and scale invariance can be made after normalization process), robustness against noise, good image reconstruction capability, and low information redundancy. Therefore, these are the most commonly used global techniques for image description in different digital image processing applications. The ORIM features are, however, unable to represent local features. Wavelet transforms (WTs), on the other hand, can effectively represent local features. WTs are not invariant to geometric transformation. We propose hybrid face recognition methods based on ORIMs and WTs. In these methods, WTs are used to represent the local features of face images and also to reduce the dimensionality of the face image which results in reducing the large computational time required for moments. WTs retain the characteristics of the original image even after reducing the size of the image. Moreover, the low frequency subband of WTs provides coefficients which are less sensitive to facial expression variation. The role of ORIMs is to capture highly discriminative rotation invariant global features with minimum redundancy. Detail experimental results show that the proposed hybrid methods based on the combination of WTs and ORIMs outperform the ORIMs methods with low processing time.

The prospective aspects of the application of digital image processing in the field of monitoring the coking process inside the boilers in thermal power plants are discussed. Timely monitoring the dynamics of the coking is an important issue to the safety of thermal power plants. Digital image processing is an effective technique to fulfill this monitoring. Edge detection and wavelet transform serve as accurate tools to investigate the ash images captured by the high-speed camera to inspect the coking. Edge detection finally presents the boundary contour of the coking zone. The advantage of wavelet transform over Fourier transform is that the former obtains not only the information on the frequency but also the information on time.

This work investigates digital image processing technology and presents a super-resolution image reconstruction approach to increase the logic of urban and rural planning spatial layout. Furthermore, this work integrates GIS technology with super-resolution image reconstruction techniques to create a digital image processing and GIS-based urban and rural spatial layout design model. In addition, this paper mathematically expresses the objects and scenes that need to be simulated as a collection of three-dimensional graphic objects stored in the computer. The platform is constructed with a three-tier system of B/S architecture, and the logical structure of the entire system is composed of the client, application server, database server, and other functional servers. Furthermore, this paper constructs the functional structure of urban and rural spatial layout planning model based on digital image processing and GIS technology and verifies the model in this paper through experimental research. The experimental statistical results verify the reasonableness of the model in this paper.

With the development of electronic technology, computer software and hardware technology, image processing and technology related to human vision, the technology of machine vision image processing has also achieved rapid development. With the rapid development of science and technology, digital image processing technology has been widely used in various fields, among which image segmentation is an extremely important technology. Image segmentation is an important intermediate technology in digital image processing. It relies on the underlying technology of image digital processing to serve high-level applications such as pattern recognition. Processing the information interaction between machine and the real world is the first step in the study of artificial intelligence, and the goal of computer vision research is to make the computer have the ability to perceive and understand image information similar to human. This paper expounds the basic principle of image segmentation technology, analyzes and discusses the methods of image segmentation, and studies the application of image segmentation technology in digital image processing.

In recent years, the level of science and technology in China has been significantly improved. Digital image processing technology is the product of the times with the rapid development of science and technology, and has been applied in many fields. Digital image processing technology includes many practical technologies, and image segmentation technology is one of them. At present, the widely used pattern recognition technology is realized by image segmentation technology. Based on this, this paper mainly studies the application of image segmentation technology in digital image processing. This paper first introduces the methods of image segmentation, including threshold segmentation, clustering segmentation and edge detection segmentation. Secondly, according to a need to extract a road image for image segmentation, and achieved very good results. Finally, this paper mainly analyzes the application of image segmentation in digital image processing. It is found that image segmentation has good application in automatic license plate recognition, biomedical engineering, remote sensing engineering and fire prevention and detection.

Due to the complexity of clinical manifestations and the lack of standardized diagnostic criteria, it is still difficult to distinguish the etiological types of congenital edentulousness corresponding to genetic defects. This paper studies the application of deep learning image processing and digital image processing in medical images in detail and analyzes the functions of congenital edentulous hotspot genes. The cases in the control group and the study group were collected, and the gene mutations of direct sequence MSX1, PAX9, AXIN2, and BMP were analyzed, and new pathogens were found. The experimental results suggest that PAX9 and MSX1 genes may have a synergistic effect in nonsyndromic congenital edentulous patients. In severely missing teeth, the role of PAX9 may be greater than that of MSX1. The experimental results will help us lay the foundation for further understanding of the disease in the future.

Image restoration is the process of obtaining a distorted/noise image and giving an approximate clear image of the original image. False focus, motion blur and noise are forms of distortion. Image restoration can be done by reversing the process called Point Extension Function (PSF). In this process, the blurred image is generated by point source imaging and can be used to restore the image lost due to the blur process. Like to form. Modern artificial intelligence (AI) applied to image processing includes facial recognition, object recognition and detection, video, image action, and visual search. It helps to develop smart applications in digital image processing.

Patch-based techniques are proven to generate promising results and outperform many of the existing state-of-art techniques for most of the applications in digital image processing. In this work we develop a patch based coherent texture synthesis technique. A patch-based anisotropic diffusion techniques combined with a novel high-frequency generating technique that can enhance line/brush strokes is also proposed. Though these techniques can be applied to many different image processing applications, here we have limited ourselves in the application to interactive digital mural painting recovery/restoration. Some empirical and practical evidence for its high quality texture synthesis and restoration for mural paintings are also presented. The experimental results show the effectiveness of the proposed algorithm.

Banana plants (Musa spp.) are one of the leading horticultural commodities in Indonesia that have high economic value and play an important role in national food security. However, banana productivity often decreases due to attacks by various diseases such as Sigatoka, Cordana, and Pestalotiopsis infections that can spread quickly. Early detection of these diseases is crucial to prevent greater losses. This study aims to develop a banana plant disease detection system based on digital image processing with the Support Vector Machine (SVM) algorithm. The research method includes the stages of banana leaf image acquisition, pre-processing using color segmentation, color and texture feature extraction, and disease type classification with the SVM algorithm. The test results show that the developed system is able to recognize banana leaf diseases with an accuracy of 97.8%, precision of 97%, and recall of 98%. These findings prove that the application of digital image processing and the SVM algorithm is effective in detecting banana plant diseases. This system is expected to be a fast, efficient, and accurate diagnostic tool for farmers to increase the productivity and quality of banana harvests.

The extraction of singly scattered photons is crucial for achieving high-quality imaging. Singular value decomposition of the optical transmission matrix (TM) can effectively separate singly and multiply scattered photons. However, traditional TM construction methods usually require capturing a large number of light field images, a process that is both time-consuming and complex, limiting its widespread use in practical applications. This paper introduces the light field contribution matrix (LFCM) model, which separates singly and multiply scattered photons using a single light field collection under low scattering conditions, enhancing measurement efficiency and enabling imaging in dynamic environments. Experimental results show that under low scattering conditions with a high proportion of singly scattered photons, the technique doubles the signal-to-noise ratio and increases the contrast by 1.5 times. In fly brain structure imaging experiments, LFCM successfully revealed hidden features and details, demonstrating its advantages in complex structural analysis. Additionally, since the technique is based on light field intensity information imaging, it has potential applications in digital image processing, particularly for reducing image blur caused by scattering, as demonstrated in the case of scatter blur correction of digital images of cataracts in the human eye. Therefore, this method provides a practical and efficient solution for high-quality imaging under low scattering conditions.

The Remote Sensing Image Scene Classification (RSI-SC) is a fundamental and challenging task in a remote sensing community area and used for earth observation application such as agricultural monitoring, civilian monitoring in military, city planning, crop planning, geomorphology, Land Use and Land Cover (LULC), metrology and soil mapping. In remote sensing image analysis, satellite and drone-captured images are used to study the Earth’s surface. The goal of image classification systems in this field is to assign meaningful labels to captured images, allowing them to be organized in a semantic structure. Such semantic ordering of images has wide applications in digital image processing and computer vision, including remote sensing, image retrieval, object recognition, image annotation, scene analysis, content-based image analysis, and video analysis. Feature extraction and representation at lower and intermediate levels was the mainstay of earlier approaches to remote sensing image analysis. The use of smaller image datasets hindered these approaches, but their great performance was demonstrated by combining numerous features and applying machine learning techniques. There has been a recent trend toward using deep learning models for remote sensing picture analysis, with hybrid deep learning approaches outperforming single models. This review article provides an in-depth examination of current deep learning methodologies, including low- and mid-level feature representation. It also provides a synopsis of image standards that are available to the public for use in remote sensing image analysis. We also compare different techniques that represent recent advances in deep learning models, addressed the challenges of remote sensing images and discussed about current applications in the field.

In this paper, we proposed a novel application based on digital image processing model in Freight Train Inspection (FTI) system. Each wagon image of freight train is captured by enterprise Automated Train Scanning System (ATSS). The main challenge of this application is automatically detecting the appropriate region of interest on the image for inspection. To address this problem, the improved adaptive multi-scaled retinex (MSR) and K-mean clustering methods are applied for image enhancement and positioning techniques. In the experimental the proposed model were found to improve the segmentation efficiency, thereby the result of inspection is optimized. The evaluation result also demonstrated the good system performance when the application is implemented using Vision Services Oriented (VSO) framework.

Handwritten digit recognition using the MNIST dataset is one of the applications in digital image processing. The selection of hyperparameters in the CNN architecture for handwriting recognition presents a challenge in achieving better recognition accuracy. This research focuses on the implementation of the DenseNet-201 architecture for recognizing handwritten digits in the MNIST dataset. The research stages include dataset preprocessing, model training, model testing, and model evaluation. The MNIST dataset consists of 60,000 training data and 10,000 testing data. Dataset preprocessing involves resizing the images to a larger size. The model training applies the DenseNet-201 architecture with selected hyperparameters such as activation functions (Softmax and ReLU), optimizers (Adam, RMSprop, and SGD), and learning rates (0.1, 0.01, and 0.001). The model testing uses one of the nine best-performing trained models. Model evaluation uses a confusion matrix to assess the accuracy and recognition performance on the MNIST dataset. The results show that the DenseNet-201 architecture with the RMSprop optimizer and a learning rate of 0.001 achieved a handwritten digit recognition accuracy of 99.49%. This study provides insights into CNN architectures and optimal hyperparameter selection for digital image processing

We present a detailed analysis of the Jacobi-Fourier moments and their applications in digital image processing. In order to reach numerical stability during the computation of the Jacobi radial polynomials a recursive approach is described. Also, some dis...

Application of digital image processing for the generation of voxels phantoms for Monte Carlo simulation