Bicubic Research Articles

Echocardiogram Image Quality Enhancement using Upsampling and Histogram Matching Methods

The prevalence of heart disease has been increasing in the last ten years. One of the cardiac diagnostic tools is echocardiography. Echocardiogram medical images provide essential information, including shape, size, pumping capacity, heart function abnormalities, and location of heart damage, but echocardiogram images have high noise content and poor contrast, as well as limitations due to differences in anatomy or body mass. This will affect the reading results of patient diagnosis. Therefore, image quality improvement is needed by removing noise and increasing image contrast. This research has improved image quality using a method with low mathematical complexity and a fast computational process. The method used is the Upsampling method to generate a reference image. The quality of the image produced was the Nearest Neighbor upsampling method: 2.8 dB, Bi-linear Interpolation: 2.78 dB, and Bi-cubic Interpolation: 2.73 dB. Furthermore, the image with the highest SNR value is processed with Histogram Matching to accelerate improving image quality. The Histogram Matching image increases quality by more than 50% with a SSIM value of 0.54. The required computational process to apply this method to each medical image has an average duration of 0.4 s. This result provides a higher value than several methods using linear scaling and speckle reducing.

Lucy Richardson and Mean Modified Wiener Filter for Construction of Super-Resolution Image

The ultimate goal of the Super-Resolution (SR) technique is to generate the High-Resolution (HR) image by combining the corresponding images with Low-Resolution (LR), which is utilized for different applications such as surveillance, remote sensing, medical diagnosis, etc. The original HR image may be corrupted due to various causes such as warping, blurring, and noise addition. SR image reconstruction methods are frequently plagued by obtrusive restorative artifacts such as noise, stair casing effect, and blurring. Thus, striking a balance between smoothness and edge retention is never easy. By enhancing the visual information and autonomous machine perception, this work presented research to improve the effectiveness of SR image reconstruction The reference image is obtained from DIV2K and BSD 100 dataset, these reference LR image is converted as composed LR image using the proposed Lucy Richardson and Modified Mean Wiener (LR-MMWF) Filters. The possessed LR image is provided as input for the stage of bicubic interpolation. Afterward, the initial HR image is obtained as output from the interpolation stage which is given as input for the SR model consisting of fidelity term to decrease residual between the projected HR image and detected LR image. At last, a model based on Bilateral Total Variation (BTV) prior is utilized to improve the stability of the HR image by refining the quality of the image. The results obtained from the performance analysis show that the proposed LR-MMW filter attained better PSNR and Structural Similarity (SSIM) than the existing filters. The results obtained from the experiments show that the proposed LR-MMW filter achieved better performance and provides a higher PSNR value of 31.65dB whereas the Filter-Net and 1D,2D CNN filter achieved PSNR values of 28.95dB and 31.63dB respectively.

Deep learning architecture for 3D image super-resolution of late gadolinium enhanced cardiac MRI.

High-resolution late gadolinium enhanced (LGE) cardiac magnetic resonance imaging (MRI) volumes are difficult to acquire due to the limitations of the maximal breath-hold time achievable by the patient. This results in anisotropic 3D volumes of the heart with high in-plane resolution, but low-through-plane resolution. Thus, we propose a 3D convolutional neural network (CNN) approach to improve the through-plane resolution of the cardiac LGE-MRI volumes. We present a 3D CNN-based framework with two branches: a super-resolution branch to learn the mapping between low-resolution and high-resolution LGE-MRI volumes, and a gradient branch that learns the mapping between the gradient map of low-resolution LGE-MRI volumes and the gradient map of high-resolution LGE-MRI volumes. The gradient branch provides structural guidance to the CNN-based super-resolution framework. To assess the performance of the proposed CNN-based framework, we train two CNN models with and without gradient guidance, namely, dense deep back-projection network (DBPN) and enhanced deep super-resolution network. We train and evaluate our method on the 2018 atrial segmentation challenge dataset. Additionally, we also evaluate these trained models on the left atrial and scar quantification and segmentation challenge 2022 dataset to assess their generalization ability. Finally, we investigate the effect of the proposed CNN-based super-resolution framework on the 3D segmentation of the left atrium (LA) from these cardiac LGE-MRI image volumes. Experimental results demonstrate that our proposed CNN method with gradient guidance consistently outperforms bicubic interpolation and the CNN models without gradient guidance. Furthermore, the segmentation results, evaluated using Dice score, obtained using the super-resolved images generated by our proposed method are superior to the segmentation results obtained using the images generated by bicubic interpolation () and the CNN models without gradient guidance (). The presented CNN-based super-resolution method with gradient guidance improves the through-plane resolution of the LGE-MRI volumes and the structure guidance provided by the gradient branch can be useful to aid the 3D segmentation of cardiac chambers, such as LA, from the 3D LGE-MRI images.

Digital rock resolution enhancement and detail recovery with multi attention neural network

High-quality digital rock images are essential for subsequent high-precision numerical simulations. But limited by the imaging capability of computed tomography (CT), high resolution digital rock images with wide imaging field of view (FOV) cannot be acquired simultaneously. To cope with this constraint, we propose a novel Multi Attention Super-Resolution Neural Network (MASR) that enhances the resolution of images with wide FOV. Considering that textures and edges are more crucial in digital rocks, MASR introduces the component attention mechanism of Component Divide-and-Conquer Super-Resolution (CDCSR) model. By redesigning the hourglass network with spatial and channel attention mechanisms, proposing a spatial attention-based mask module, and optimizing the component attention mask calculation process, MASR delivers higher information utilization with fewer parameters and faster training than CDCSR. In addition, we optimize the depth of MASR to trade off speed and super-resolution quality. Furthermore, we retrain several state-of-the-art models. Through quantitative evaluations and qualitative visualizations, it is verified that MASR can recover sharper edges while removing noise, and obtain digital rock images with superior quality and reliability. The pixel-wise relative errors of MASR reconstructions are reduced by 15%–26% over bicubic interpolation method. Our codes and pretrained model are publicly available at https://github.com/MHDXing/MASR-for-Digital-Rock-Images.

A Terahertz Fast Imaging Method for Debonding Defects of Thermal Barrier Coatings Based on Dual-Channel Convolutional Neural Network

To tackle the inefficiency of terahertz (THz)-based C-scan defect detection for thermal barrier coatings (TBCs), a dual-channel convolutional neural network–based THz fast imaging method is proposed. In this paper, the finite-difference time-domain (FDTD) method is used to prepare the training set. In the numerical simulation, the actual C-scan step is simulated by grid division of different sizes. The large step THz image is preliminarily reconstructed by bicubic interpolation, and then the deep and shallow features in the image are extracted by the dual-channel convolution neural network and the image under small step is reconstructed by different weight refusion, so as to improve the detection efficiency by reducing the number of C-scan points. Gaussian white noise with different distributions is employed when simulating the real test image. The experimental results show that compared with bicubic, ICBI, SRCNN, and ResNet, the dual-channel convolutional neural network improves PSNR (peak signal-to-noise ratio) by 2.85, 2.81, 2.25, and 1.54, and improves by 0.019, 0.014, 0.014, and 0.009 on SSIM (structural similarity).

Historical Text Image Enhancement Using Image Scaling and Generative Adversarial Networks.

Historical documents such as newspapers, invoices, contract papers are often difficult to read due to degraded text quality. These documents may be damaged or degraded due to a variety of factors such as aging, distortion, stamps, watermarks, ink stains, and so on. Text image enhancement is essential for several document recognition and analysis tasks. In this era of technology, it is important to enhance these degraded text documents for proper use. To address these issues, a new bi-cubic interpolation of Lifting Wavelet Transform (LWT) and Stationary Wavelet Transform (SWT) is proposed to enhance image resolution. Then a generative adversarial network (GAN) is used to extract the spectral and spatial features in historical text images. The proposed method consists of two parts. In the first part, the transformation method is used to de-noise and de-blur the images, and to increase the resolution effects, whereas in the second part, the GAN architecture is used to fuse the original and the resulting image obtained from part one in order to improve the spectral and spatial features of a historical text image. Experiment results show that the proposed model outperforms the current deep learning methods.

High-resolution aeromagnetic map through Adapted-SRGAN: A case study in Québec, Canada

Due to their cost-effectiveness, aeromagnetic data have been acquired for decades to guide mineral exploration. Aeromagnetic map enhancement is immensely useful as it allows dykes, faults, and other geological structures to be highlighted more clearly, therefore assisting the geologist with a better understanding of the geological process. Over the last years, technological improvements allowed increasing the sensitivity of airborne magnetic data acquisition systems and the accuracy of navigation instrumentation, which resulted in higher resolution (HR) maps. Such higher resolution implies close lines of flight, which typically implies smaller area coverage. On the other hand, the vintage aeromagnetic surveys have high coverage but a much lower resolution. Geological interpretation for regions where only low-resolution aeromagnetic data (LR) are available remains challenging for geologists. Hence, we adapted and trained a deep neural network to address this problem by learning the statistical relations between collocated LR and HR airborne magnetic data. First, we trained a Super-Resolution (SR) network based on an Adapted-SRGAN (ASRGAN) in the source (training) set for mapping LR images to HR images. Second, given this trained network, we generate HR images in the target (test) set where only LR images are available, with a 4 × times resolution increase. We validated the generalization of our model using aeromagnetic maps from several regions of the Québec province, by computing Peak Signal to Noise Ratio (PSNR) and Structural Similarity index (SSIM) between super-resolved GAN outputs with ground truth HR aeromagnetic images. The performance of the proposed approach is compared to bicubic interpolation and conventional SRGAN, with a slight improvement in terms of the SSIM and PSNR but with about 35 percent reduction in the computational time required to train the network.

Super-resolution for terrain modeling using deep learning in high mountain Asia

High Mountain Asia (HMA) is characterized by some of the most complex and rugged terrain conditions in the world. However, high resolution terrain data are not easy to quickly acquire from the area due to difficulties in accessing the region. In this study, we trained a modified super-resolution residual network (MSRResNet) to develop super-resolution (SR) digital elevation models (DEMs) in the HMA areas by using freely available DEM data from the HMA region and limited high resolution (HR) DEMs from other areas to train the model. In this network, a new loss function was constructed that considered the terrain parameters of slope and curvature to constrain the network learning and convergence. The proposed method was applied to and validated by data from the Hengduan Mountains in the southeastern part of HMA, which is a world-famous longitudinal belt of mountains and canyons. A comparative analysis between the current and existing methods (i.e., SRGAN and Bicubic interpolation) was conducted to assess the effectiveness of the proposed approach. The experimental results were also investigated and evaluated by visual inspection and analysis of the terrain parameters. The results demonstrate that the proposed MSRResNet super-resolution process can achieve highly accurate terrain data by downscaling DEMs in HMA. This SR process also outperforms the other comparable methods. Compared to the Bicubic interpolation method, the RMSE and MAE accuracy are improved by 32.17% and 33.97%, and compared to the SRGAN method, the RMSE and MAE accuracy are improved by 39.15% and 32.47%. The HR DEM generated by the new method is more conducive to improving the accuracy of extracted terrain features, such as stream networks. It is promising to apply this model on other areas on Earth or even other planets with terrain similar to that of HMA.

One‐kilometre monthly air temperature and precipitation product over the Mongolian Plateau for 1950–2020

AbstractThe fragile ecological environment of the Mongolian Plateau (MP) is sensitive to climate change. It is necessary to fully understand the temperature and precipitation changes on the MP to ensure regional sustainable development. Most existing studies are conducted based on station observations which suffer from sparse distribution and limited spatial representativeness and cannot perfectly represent the climatic conditions of the whole region. By contrast, the long‐term, spatially continuous reanalysis products offer new opportunities for MP climate change research. However, the coarse spatial resolution and large uncertainties of reanalysis data limit their applicability to provide reliable climate information at finer scales. Within the delta downscaling framework, we downscale and correct the state‐of‐the‐art European Centre for Medium‐range Weather Forecasts ReAnalysis 5 land portion (ERA5‐Land) using a high‐resolution WorldClim reference climatology and generate a long‐term (1950–2020) 1‐km monthly air temperature and precipitation downscaled dataset. During the downscaling process, the results obtained by nearest neighbour, bilinear and bicubic interpolation methods are evaluated and compared. Air temperature result of bilinear method and precipitation result of nearest neighbour method, which have the best accuracy, are taken as the final downscaled data. The evaluation shows that the downscaled data has acceptable accuracy (overall MBE = −0.41°C, Corr = 0.997, RMSE = 1.51°C for air temperature, and MBE = 1.58 mm·month−1, Corr = 0.747, RMSE = 24.24 mm·month−1 for precipitation), and outperforms the original ERA5‐Land and widely used Climatic Research Unit (CRU) data. In addition, the downscaled data can also provide accurate and detailed temperature and precipitation change trends over the MP. With fine spatial resolution, long‐time span and good spatial continuity, the downscaled air temperature and precipitation will be a useful data source to explore climate change over the MP.

Novel hybrid computed axial phono-cardiac tomography system for heart sound sources localization

According to recent reports by the world health organization (WHO), heart valve disease is one of the leading causes of death around the world. Due to the complex anatomy of the heart, it is quite challenging task to develop a system for heart sound localization, and therefore, there is a need for an accurate axial heart sound localization and tomography imaging for detecting heart valve disease. The proposed approach of Axial heart sound tomography aims to improve the accuracy and speed of heart-valves disease diagnostics. We have developed a simulation hybrid system of a chest shape microphone array with a CT image at the same chest level for functional heart sound tomography imaging that has been proposed, analyzed and computationally evaluated. In the proposed method, the heart valve sound distribution responses are collected simultaneously from the array of surrounding microphones. To increase the response array resolution, an upsampling, using bicubic interpolation, is performed to provide, mathematically, a higher sound localization resolution in the process of axial heart sound tomographic image reconstruction. The sound localization simulation of two different scenarios are considered, first for aortic and pulmonary valves and second for bicuspid and tricuspid valves. The results show that the proposed approach can reconstruct a heart-valves sound image with a resolution of 15.32 mm and 11.9 mm in x- and y-directions respectively. The proposed simulation hybrid system shows its ability to accurately localize the heart valves with high resolution in both × and y directions. Thus, the proposed method provides a useful analytical and diagnostic tool for heart valves by estimating their sound-based problems.

A Study on the Super Resolution Combining Spatial Attention and Channel Attention

Existing CNN-based super resolution methods have low emphasis on high-frequency features, resulting in poor performance for contours and textures. To solve this problem, this paper proposes single image super resolution using an attention mechanism that emphasizes high-frequency features and a feature extraction process with different depths. In order to emphasize the high-frequency features of the channel and space, it is composed of CSBlock that combines channel attention and spatial attention. Attention block using 10 CSBlocks was used for high-frequency feature extraction. In order to extract various features with different degrees of feature emphasis from insufficient low-resolution features, features were extracted from structures connected with different numbers of attention blocks. The extracted features were expanded through sub-pixel convolution to create super resolution images, and learning was performed through L1 loss. Compared to the existing deep learning method, it showed improved results in several high-frequency features such as small object outlines and line patterns. In PSNR and SSIM, it showed about 11% to 26% improvement over the existing Bicubic interpolation and about 1 to 2% improvement over VDSR and EDSR.

Respiration and heartbeat motion correction of intraoperative thermographic images in neurosurgery

In this paper, a new respiration and heartbeat motion artifacts correction method for thermographic images in neurosurgery is presented. Aims to solve the ambiguity of the respiration and heartbeat motion artifacts correction of thermographic images in neurosurgery. The proposed method consists of two main steps; firstly, a complex steerable pyramid is employed to separate the local wavelets of the phase for each thermographic image. Thereafter, temporal filtering is carried out to pass respiration and heartbeat frequencies and reject other frequencies outside that range followed by image reconstruction. Secondly, the optimized and adjusted-combined local and global (OA-CLG) method is applied to detect respiration and heartbeat motion artifacts only. Afterwards, a bicubic interpolation is carried out to compensate the estimated motion artifacts. To evaluate the proposed method’s performance and accuracy, 10 clinical thermographic datasets were examined. Results show that with the proposed method respiration and heartbeat motion artifacts can be corrected while preserving the image structures, properties, spatial resolution, and maintaining temperature values.

Super-Resolution of Dental Panoramic Radiographs Using Deep Learning: A Pilot Study.

Using super-resolution (SR) algorithms, an image with a low resolution can be converted into a high-quality image. Our objective was to compare deep learning-based SR models to a conventional approach for improving the resolution of dental panoramic radiographs. A total of 888 dental panoramic radiographs were obtained. Our study involved five state-of-the-art deep learning-based SR approaches, including SR convolutional neural networks (SRCNN), SR generative adversarial network (SRGAN), U-Net, Swin for image restoration (SwinIr), and local texture estimator (LTE). Their results were compared with one another and with conventional bicubic interpolation. The performance of each model was evaluated using the metrics of mean squared error (MSE), peak signal-to-noise ratio (PNSR), structural similarity index (SSIM), and mean opinion score by four experts (MOS). Among all the models evaluated, the LTE model presented the highest performance, with MSE, SSIM, PSNR, and MOS results of 7.42 ± 0.44, 39.74 ± 0.17, 0.919 ± 0.003, and 3.59 ± 0.54, respectively. Additionally, compared with low-resolution images, the output of all the used approaches showed significant improvements in MOS evaluation. A significant enhancement in the quality of panoramic radiographs can be achieved by SR. The LTE model outperformed the other models.

Super-Resolution Techniques in Photogrammetric 3D Reconstruction from Close-Range UAV Imagery

Current Multi-View Stereo (MVS) algorithms are tools for high-quality 3D model reconstruction, strongly depending on image spatial resolution. In this context, the combination of image Super-Resolution (SR) with image-based 3D reconstruction is turning into an interesting research topic in photogrammetry, around which however only a few works have been reported so far in the literature. Here, a thorough study is carried out on various state-of-the-art image SR techniques to evaluate the suitability of such an approach in terms of its inclusion in the 3D reconstruction process. Deep-learning techniques are tested here on a UAV image dataset, while the MVS task is then performed via the Agisoft Metashape photogrammetric tool. The data under experimentation are oblique cultural heritage imagery. According to results, point clouds from low-resolution images present quality inferior to those from upsampled high-resolution ones. The SR techniques HAT and DRLN outperform bicubic interpolation, yielding high precision/recall scores for the differences of reconstructed 3D point clouds from the reference surface. The current study indicates spatial image resolution increased by SR techniques may indeed be advantageous for state-of-the art photogrammetric 3D reconstruction.

Efficient Windows malware identification and classification scheme for plant protection information systems.

Due to developments in science and technology, the field of plant protection and the information industry have become increasingly integrated, which has resulted in the creation of plant protection information systems. Plant protection information systems have modernized how pest levels are monitored and improved overall control capabilities. They also provide data to support crop pest monitoring and early warnings and promote the sustainable development of plant protection networks, visualization, and digitization. However, cybercriminals use technologies such as code reuse and automation to generate malware variants, resulting in continuous attacks on plant protection information terminals. Therefore, effective identification of rapidly growing malware and its variants has become critical. Recent studies have shown that malware and its variants can be effectively identified and classified using convolutional neural networks (CNNs) to analyze the similarity between malware binary images. However, the malware images generated by such schemes have the problem of image size imbalance, which affects the accuracy of malware classification. In order to solve the above problems, this paper proposes a malware identification and classification scheme based on bicubic interpolation to improve the security of a plant protection information terminal system. We used the bicubic interpolation algorithm to reconstruct the generated malware images to solve the problem of image size imbalance. We used the Cycle-GAN model for data augmentation to balance the number of samples among malware families and build an efficient malware classification model based on CNNs to improve the malware identification and classification performance of the system. Experimental results show that the system can significantly improve malware classification efficiency. The accuracy of RGB and gray images generated by the Microsoft Malware Classification Challenge Dataset (BIG2015) can reach 99.76% and 99.62%, respectively.

Image Interpolation Based on Spiking Neural Network Model

Image interpolation is used in many areas of image processing. It is seen that many techniques developed to date have been successful in both protecting edges and increasing image quality. However, these techniques generally detect edges with gradient-based linear calculations. In this study, spiking neural networks (SNNs), which are known to successfully simulate the human visual system (HVS), are used to detect edge pixels instead of the gradient. With the help of the proposed SNN-based model, the pixels marked as edges are interpolated with a 1D directional filter. For the remaining pixels, the standard bicubic interpolation technique is used. Additionally, the success of the proposed method is compared to known methods using various metrics. The experimental results show that the proposed method is more successful than the other methods.

Evaluation of near-surface air temperature reanalysis datasets and downscaling with machine learning based Random Forest method for complex terrain of Turkey

Near-surface air temperature is a key variable used in a wide range of applications showing the environmental conditions of the Earth. Standard meteorological observations generally provide the best estimation with high accuracy over time for a small area of influence. However, considerable uncertainty arises when point measurements are extrapolated or interpolated over much larger areas. Satellite and reanalysis products have emerged as a viable alternative or supplement to in situ observations due to their availability over vast ungauged regions. Thus, spatial patterns of air temperature can be derived from these products.In this study, we evaluate the performance of several reanalysis products of near-surface air temperature to determine the best product in estimating daily and monthly air temperatures across the complex terrain of Turkey. European Center for Medium-Range Weather Forecasts Reanalysis version 5 (ERA5), (AgERA5), (ERA5-Land), the Modern-Era Retrospective Analysis for Research and Applications version 2 (MERRA2), and the Japanese 55-year Reanalysis (JRA-55) products are assessed with 1120 ground-based gauge stations for the period 2015–2019 over complex terrain having different climate classes according to Köppen-Geiger classification scheme and land surface types. Moreover, several traditional and more sophisticated machine learning downscaling algorithms are applied to increase the products’ spatial resolution. The agreement between ground observations and the different products and the downscaled temperature product is investigated by using a set of commonly employed statistical estimators of mean absolute error (MAE), correlation coefficient (CC), root-mean-square error (RMSE), and bias.Performance analysis of reanalysis air temperature products with ground-based observations on daily and monthly time series has shown that among the five datasets, the AgERA5 product is superior in estimating air temperature over Turkey, both seasonally and annually. Compared to the other datasets, ERA5 and ERA5-Land show similar performance and reach the second highest performance score at both daily and monthly time steps. In terms of improving product performance with spatial downscaling, among the distance-based methods, the best overall performance is obtained by bicubic interpolation with a slight increase in the product performance in monthly and daily time series. However, depending on the season, the Random Forest algorithm's performance is far superior to all other methods used in this study.

Feature Pyramid Networks and Long Short-Term Memory for EEG Feature Map-Based Emotion Recognition.

The original EEG data collected are the 1D sequence, which ignores spatial topology information; Feature Pyramid Networks (FPN) is better at small dimension target detection and insufficient feature extraction in the scale transformation than CNN. We propose a method of FPN and Long Short-Term Memory (FPN-LSTM) for EEG feature map-based emotion recognition. According to the spatial arrangement of brain electrodes, the Azimuth Equidistant Projection (AEP) is employed to generate the 2D EEG map, which preserves the spatial topology information; then, the average power, variance power, and standard deviation power of three frequency bands (α, β, and γ) are extracted as the feature data for the EEG feature map. BiCubic interpolation is employed to interpolate the blank pixel among the electrodes; the three frequency bands EEG feature maps are used as the G, R, and B channels to generate EEG feature maps. Then, we put forward the idea of distributing the weight proportion for channels, assign large weight to strong emotion correlation channels (AF3, F3, F7, FC5, and T7), and assign small weight to the others; the proposed FPN-LSTM is used on EEG feature maps for emotion recognition. The experiment results show that the proposed method can achieve Value and Arousal recognition rates of 90.05% and 90.84%, respectively.

Deep learning based image reconstruction at any speeds for faster pavement texture measurement using 3D laser technology

ABSTRACT Recently, the super-resolution network PT-SRGAN was developed for faster pavement texture measurement using 3D laser technology at 0.1 mm resolution from only six predetermined faster speeds. This paper further introduces an extended application of PT-SRGAN in combination with bicubic interpolation to enhance the low-resolution 3D pavement data collected at any speeds for more flexible practices. The research team collected two datasets on ten field pavement sections: (1) Dataset-1 was the true 0.1 mm 3D texture data collected at very low speeds (<1.5 mph); and (2) Dataset-2 was the low-resolution 3D texture data collected at three faster speeds (7.5, 15, and 30 mph). The efficacy of the extended PT-SRGAN was validated by reconstructing 0.1 mm data from manually downsized low-resolution Dataset-1 and real low-resolution Dataset-2. First, the superior performance of the proposed method was demonstrated by examining two evaluation metrics calculated between ground truth and reconstructed images at different speeds using Dataset-1. Further, seven 3D areal texture parameters were calculated and averaged for reconstructed Dataset-2, and compared with those of Dataset-1 to demonstrate that the proposed method shows promising performance to enhance real low-resolution 3D texture images to 0.1 mm data for any faster pavement texture evaluation.

Research on Image Measurement Method of Flat Parts Based on the Adaptive Chord Inclination Angle Algorithm

To accurately measure the critical dimensions of flat parts using machine vision, an inspection method based on the adaptive chord inclination angle progressively screening the segmentation points of graphic elements was proposed in this study. The method doubled the size of the part image using bicubic interpolation, extracted the single-pixel contour with more detailed information, and designed an adaptive step size to obtain the front and back chord inclination angles of the contour. The method of complementing the front and back chord inclination angles was employed to avoid the negative effects of contour jaggedness, thereby obtaining the contour segmentation points after the initial screening. The segmentation points obtained in the initial sieving were divided into different point clusters according to the distance, and the contour, which was segmented by two segmentation points in different point clusters, was fitted using the least squares. The fitting results were evaluated, and all the fitting results were selected using the improved non-maximum suppression (NMS) algorithm to obtain the precisely selected segmentation points of the graphic elements. Consequently, the segmented individual graphic elements were fitted with the segmentation points as constraints to obtain the key dimensions of the closed part. The developed method could accurately find the contour segmentation points, and the relative error was less than 0.6%.