Real-time Digital Image Processing Research Articles

Enhancing Tower Crane Safety: A UAV-Based Intelligent Inspection Approach

Tower cranes play a crucial role in the construction industry, facilitating the vertical and horizontal movement of materials and aiding in building construction, especially for high-rise structures. However, tower crane accidents can lead to severe consequences, highlighting the importance of effective safety management and inspection. This paper presents a novel approach to tower crane safety inspections using Unmanned Aerial Vehicles (UAVs) equipped with high-definition cameras and an intelligent inspection APP system. The system utilizes real-time kinematic (RTK) positioning and digital image processing to perform efficient and comprehensive inspections, reducing the reliance on manual labor and associated risks. A case study demonstrated the method’s practicality and effectiveness, with the UAV inspections capable of identifying 76.3% of major hazards, 64.8% of significant hazards, and 76.2% of general hazards within a 30-minute timeframe. Preliminary identification rates were also promising. Despite the initial requirement for manual drone piloting and the current manual review of images, the approach shows significant potential for enhancing safety in the construction industry. Future work will focus on integrating AI for hazard recognition and automating the inspection process further. The proposed method marks a step forward in tower crane safety management, offering a more efficient and accurate alternative to traditional inspection methods.

Design of real-time digital image processing system based on high performance computing

Design of real-time digital image processing system based on high performance computing

Clinical performance of a smartphone-based low vision aid

Real-time digital image processing to optimally enhance low vision is now realizable with recent advances in personal computers. This study aimed to evaluate the efficacy of a wearable smartphone-based low vision aid (LVA) with customizable vision enhancement in patients with visual impairment. We recruited 35 subjects with visual impairment and who were literate and cognitively capable. The subjects completed a training session and were provided a smartphone-based LVA for a 4-week use. Visual functions including binocular best-corrected distance, intermediate, and near visual acuities; reading performance (reading speed and accuracy); and facial recognition performance were measured at baseline and after 4-weeks use. All subjects also completed the Low Vision Quality of Life (LVQOL) Questionnaire. Thirty-four subjects (mean age, 43.82 ± 15.06 years) completed the study. Significant improvements in binocular best-corrected distance, intermediate, and near visual acuities were observed after smartphone-based LVA use (all p < 0.001). Reading accuracy and facial recognition performance also improved significantly (p = 0.009 and p < 0.001, respectively), but reading speed did not. LVQOL scores significantly improved after 4 weeks of use in subjects aged < 40 years (p = 0.024), but not in subjects aged ≥ 40 years (p = 0.653). Ocular and non-ocular adverse events were infrequent and resolved when the device was removed. The smartphone-based LVA with customizable vision enhancement could provide clinically significant improvements in the visual function of patients with visual impairment and was generally well tolerated. This study suggests that the smartphone-based LVA would be beneficial for visual rehabilitation.

Digital image processing systems based on functional-oriented processors with a homogeneous structure

Real-time digital image processing as part of mobile systems requires solving the problems of development high-performance specialized processors for intraframe image processing. The graphs of intraframe processing algorithms are analysed and their parameters are evaluated, primarily the degree of parallelism. The architectures of functional-oriented processors based on homogeneous computing environments for ultrafast intraframe processing are described. The results of research and development of microelectronic implementation of such processors are presented.

Hardware implementation and validation of a traffic road sign detection and identification system

Reconfigurability and parallel computing capability of field programmable gate array (FPGA) devices are highly exploited in real-time digital image and video processing applications. In this field, real-time traffic road signs detection systems present a huge interest since they help to assist drivers and decrease accidents. In this paper, we propose an FPGA-based hardware implementation of road signs detection and identification system. The proposed system can achieve real-time video constraints while assuring a high-level accuracy in terms of detection rate. The performance of the system in terms of processing latency was evaluated relatively to the reaction distance, the braking distance and the vehicle speed. The evaluation results show that our system can support real-time driving conditions until the speed of 110 km/h. To prove the validity of the proposed implementation, a hardware co-simulation strategy was applied. This is based on the use of Matlab/Xilinx system generator. A comparison of the co-simulation results shows the effectiveness of the developed architecture.

Bit-Level Systolic Architecture for a Matrix-Matrix Multiplier

Highly efficient arithmetic operations are necessary to achieve the desired performance in many real-time systems and digital image processing applications. In all these applications, one of the important arithmetic operations frequently performed is to multiply and accumulate with small computational time. In this paper, a 4-bit serial - parallel multiplier, which can perform both positive and negative multiplications, is presented. Baugh-Wooley algorithm necessitates complementation of last bit of each partial product except the last partial product in which all but the last bit are complemented. In the proposed algorithm all bits of the last partial product are complemented. This modification results in considerable reduction in hardware compared to Baugh-Wooley multiplier. This multiplier can be used for implementation of discrete orthogonal transforms, which are used in many applications, including image and signal processing. This paper presents a 2D bit-level systolic architecture for a matrixmatrix multiplier. A comparison with similar structures has shown that the proposed structure performs better.

Performance Evaluation of Squaring Operation by Vedic Mathematics

Highly efficient arithmetic operations are necessary to achieve the desired performance in many real-time systems and digital image processing applications. In all these applications, one of the important arithmetic operations frequently performed is to multiply and accumulate with a small computational time (delay). As in all the arithmetic operations, it is the squaring which is most important in finding the transforms or the inverse transforms in signal processing. The squaring operation occupies most of the computing time; therefore, it becomes imperative to concentrate on improving the speed with which we square the numbers. The squaring operation also forms the backbone in cryptography. In this paper, we propose a technique for implementing squaring operation using Vedic methods in VHDL and evaluate the performance. The results are compared with the conventional Booth’s algorithm in terms of time delay and area occupied on the Xilinx Virtex 4vlx15sf363-12.

Focus on lasers and imaging

Focus on lasers and imaging

A Remote Laboratory for Real-Time Digital Image Processing on Embedded Systems

The purpose of this paper is to present a Remote Laboratory on embedded systems focused in real-time digital image processing. This laboratory consists of a Main Web Server and several Workstations which are designed for digital image retrieval from a CMOS Image Sensor and real-time image processing on a Digital Signal Processor development platform. The Main Web Server redirects the authorised remote users to available Workstations in order to execute and verify their image processing algorithms or test their system designs using a developed Application Programming Interface. Through user-friendly web pages users can interfere with the hardware parameters and observe the results of their solutions.

Application development with the FlexWAFE real-time stream processing architecture for FPGAs

The challenges posed by complex real-time digital image processing at high resolutions cannot be met by current state-of-the-art general-purpose or DSP processors, due to the lack of processing power. On the other hand, large arrays of FPGA-based accelerators are too inefficient to cover the needs of cost sensitive professional markets. We present a new architecture composed of a network of configurable flexible weakly programmable processing elements, Flexible Weakly programmable Advanced Film Engine (FlexWAFE). This architecture delivers both programmability and high efficiency when implemented on an FPGA basis. We demonstrate these claims using a professional next-generation noise reducer with more than 170G image operations/s at 80% FPGA area utilization on four Virtex II-Pro FPGAs. This article will focus on the FlexWAFE architecture principle and implementation on a PCI-Express board.

Distortion correction for a wide-angle lens based on real-time digital image processing

Images captured with wide-angle lenses suffer from spatial distortion, which prevents accurate measure and feature extraction. In this work, a mathematical model based on polynomial mapping is used to map the images from distorted image space onto the corrected image space. The model parameters include the polynomial coefficients, distortion center, and dot center. A new technique to estimate the distortion center based on the evaluation function is presented. In dot center estimation, a seed fill algorithm is used to traverse all pixels. The expansion polynomial was obtained by using cubic spline interpolation. With a field programmable gate array (FPGA), the real-time distortion correction is implemented, which is independent of a computer. The major benefit using FPGA is that the same circuit can be used for other circularly symmetric wide-angle lenses without being modified.

Improved two-dimensional product imaging: The real-time ion-counting method

A novel ion-counting method for significantly improving the spatial resolution and detection sensitivity of two-dimensional product imaging in molecular beam experiments is presented. The method makes use of real-time digital image processing to retrieve, threshold, and determine the local maximum of each ion hitting a microchannel plate assembly. The current version can process data at rates up to 3.07 Mbyte/s, and methods for accelerating this rate are proposed.

Contrast variations in white-light speckle interferometry with application to 3D profilometry

We investigated the interference of a white-light speckle pattern, obtained by reflection from a rough surface, with a reference beam. A theoretical analysis has been developed to derive the expression of the speckle contrast, observed in the image plane of a telescopic system, as a function of all the parameters of the interference device. A numerical evaluation of the contrast dependence on the optical path difference is presented, as well as its experimental verification. On the basis of these results on the speckle contrast properties a method has been developed for 3D profilometry of relatively large optically rough objects, using real-time digital image processing.

Advances in the development of the interferometric otoscope.

Present measurement techniques for middle ear function have inherent limitations because they are either spatially insensitive (acoustic immittance) or descriptive and qualitative in nature (otoscopy). By integrating advances in electrooptic technology (fiber optics, miniature diode lasers, solid-state detector arrays) and digital processing, further advances are possible. On the basis of measurements taken with electronic speckle-pattern interferometry on human temporal bones and models, we demonstrate quantitative static and dynamic vibration/displacement characteristics of the tympanic membrane with high spatial resolution. Our presentation emphasizes advantages of optically based methods and demonstrates computerized signal processing capable of fringe localization, enhancement, and counting. Miniaturization and real-time digital image processing in the clinical setting is the goal of this research.

Impulse noise correction in TV transmission

We describe a real-time digital image processing approach for correcting the impulse noise impairment in television transmission. The described algorithm identifies the pixels impaired with impulse noise and subsequently corrects them. A real-time processing board is constructed to demonstrate the algorithm. >

A low-cost system for digital image processing of moving images in real time: application to X-ray fluoroscopy

A low-cost image processing system and two real-time digital image processing algorithms using parallel processing have been developed to provide automatic equalization of moving images at video frequencies. The algorithms differ in the processing time required and the characteristics of the final image. The system takes an entire image from the video sequence, digitizes it, automatically calculates its histogram and the lowest and highest useful grey levels, equalizes the histogram, and finally converts it to the analogue image. Methods were applied to X-ray fluoroscopy (angiography) images, which are usually saturated or of uniform intensity (low contrast), leading to information loss. An application of these methods is in interventional radiography, which requires images in real time.

Performance characteristics of a real-time digital x-ray fluoroscopic system using an intensified charge injection device camera and CsI:Na crystal.

A digital x-ray photoelectronic imaging system has been constructed using an optically flat 152-mm-diam, 2.5-mm-thick CsI:Na scintillating crystal. X-ray images formed by the scintillator are viewed by a single microchannel plate intensified charge injection device (ICID) camera and digitized at a real-time rate by a computerized frame-grabbing system. Video images are recorded and selected image frames are subjected to image processing and analysis schemes. Parameters governing the performance characteristics of the system are determined accordingly. For a 152-mm field size at the crystal plane, a spatial resolution limit of 1.50 +/- 0.10 1p/mm (1 sigma) measured at the 4% level of the modulation transfer function of the system has been obtained. This result is consistent with the measurements of the full width at half-maximum of the line spread function which is found to be 645 +/- 35 microns (1 sigma). Similarly, the intrinsic resolution of the CsI:Na scintillator only was found to be 6.5 +/- 0.5 1p/mm (1 sigma). Contrast ratio measurements, which are mainly determined by the quality of phosphor, have indicated a value of 12.1 +/- 0.6, whereas minimum visible details are observed at radiation exposure rates of 100 microR/s. This limit has been reduced to 10 microR/s using the single-scan integrating option provided by the ICID camera. A maximum contrast resolution of 1% corresponding to 100 statistically significant meaningful gray levels is achieved at a maximum exposure rate of 1000 microR/s. Consequently, although the imaging capability of the present system compares favorably with that of conventional x-ray video-fluoroscopic systems, a better performance may be achieved by using a higher resolution cooled dual-microchannel intensified CID camera in conjunction with a thinner CsI:Na crystal and a real-time digital image processing subsystem.

High-resolution Kerr observation of magnetic domains

A high-resolution Kerr microscope was developed for observations of in-plane or perpendicular magnetized domains with widths down to 0.2 μm. Our main interest was to optimize the magnetization contrast, derived from the Kerr rotation and Kerr ellipticity, without anti-reflective coating of the magnetic surface. This is made possible by parallel argon laser illumination, a quarter-wave retarder, an imaging aperture matched to the resolution necessary and by optimized adjustment of the illumination and polarizing optics, controlled by a contrast-measuring difference photo-amplifier. The effect of laser interference patterns is minimized by a wobble plate which rotates in the beam. Real-time digital video image processing (512 × 512 pixels × 16 bit) allows subtraction of non-magnetic background and further contrast enhancement. Wall displacements are observed with a time resolution of 50 half-frames per second. Domains and their response to applied fields were investigated in magnetic microstructures.

Digital Image Enhancement and Noise Filtering by Use of Local Statistics

Computational techniques involving contrast enhancement and noise filtering on two-dimensional image arrays are developed based on their local mean and variance. These algorithms are nonrecursive and do not require the use of any kind of transform. They share the same characteristics in that each pixel is processed independently. Consequently, this approach has an obvious advantage when used in real-time digital image processing applications and where a parallel processor can be used. For both the additive and multiplicative cases, the a priori mean and variance of each pixel is derived from its local mean and variance. Then, the minimum mean-square error estimator in its simplest form is applied to obtain the noise filtering algorithms. For multiplicative noise a statistical optimal linear approximation is made. Experimental results show that such an assumption yields a very effective filtering algorithm. Examples on images containing 256 × 256 pixels are given. Results show that in most cases the techniques developed in this paper are readily adaptable to real-time image processing.