Search

Currently, a reliable and easy-to-use method to monitor flocculation in the water treatment process is highly demanded, especially for small water purification stations. For this problem, in situ images were used to analyze the flocculation process under different conditions via jar tests. A texture feature of the gray level co-occurrence matrix was found to be helpful for monitoring the floc status, such as growth rate and settling velocity. To further verify this finding, we established the correlation between the texture time sequence curve (TTSC) and its corresponding floc status. The slope of the TTSC during the growth phase and during the settling phase can describe the growth rate and the settling velocity, respectively, i.e., the higher the slope, the higher the growth rate and settling velocity. In addition, significant differences between the TTSCs in various abnormal conditions and the normal condition of coagulation can be identified. By using the TTSC for detecting abnormal conditions, we again verified that the texture feature can reliably reflect the flocculation process. Our study helps to develop a low-cost, stable, and simple method for monitoring flocculation and detecting abnormal conditions, which can effectively be used in the operation and management of water treatment plants. HIGHLIGHTS Compared with the morphological features, the texture features are more stable.; The texture time sequence curve (TTSC) can describe flocculation.; Variation characteristics of TTSC can detect abnormal coagulation conditions.; TTSC-based methods can be readily applied online.;

A novel method of disease diagnosis, based on images that capture every part of a diseased plant, such as the leaf, the fruit, the root, etc., is presented in this paper. As is well known, the plant genotypic and phenotypic characteristics can significantly impact how plants are affected by viruses, bacteria, or fungi that cause disease. Assume that these data are unknown at the outset and that the appropriate precautions are not taken to prevent classifications skewed toward uninteresting traits. An approach to avoid categorization bias brought on by the morphology of leaves is suggested in this study. The basis of this approach is the extraction of textural features. Additionally, Bayesian Optimization is suggested to obtain training hyperparameters that enable the creation of better-trained artificial neural networks. First, we initially pre-processed the images from the PlantVillage dataset to remove background noise. Then, tiles from images were used to reduce any potential bias from leaf form. Finally, several cutting-edge tiny convolutional neural networks (CNNs), created for contexts with little processing power, were trained on a new dataset of 85 × 85 × 3 px images. MobileNet, which had a 96.31% accuracy rate, and SqueezeNet, which had a 95.05% accuracy rate, were the models that predicted the best performance. The results were then examined using Precision and Recall measures, which are important for identifying plant diseases.

In land-use classification of hyperspectral remote sensing (RS) images, traditional classification methods often experience large amount of datasets and low efficiency. To solve these problems, a fast machine-learning method, the extreme learning machine (ELM) algorithm, was introduced. However, basic use of the ELM usually encounters problems of unstable classification results and low classification accuracy. Hence, in this paper, optimization methods for ELM-based RS image classification were mainly discussed and applied to solve the bottleneck problems. From the three perspectives of ensemble learning, making full use of image texture features, and deep learning, three classification optimization methods have been designed and implemented. The results show that: 1) To some extent, all the three methods can achieve a balance between classification accuracy and efficiency, i.e., they can maintain the advantage of ELM algorithm in classification efficiency and speed while have better classification accuracy; 2) The image texture feature optimization method (LBP-KELM) solves the problem of unsatisfactory classification results experienced by the ensemble learning optimization method (Ensemble-ELM) and further improves classification accuracy. However, the classification results are sensitive to the type of dataset; and 3) Fortunately, the optimization method combined with deep learning (CNN-ELM) can meet the application needs of multiple datasets. Furthermore, it can also further improve classification accuracy.

Nowadays, diagnosis of thyroid nodules is mainly based on clinical methods, which requires a lot of manpower and medical resources. Therefore, this work proposes an automated thyroid ultrasound nodule diagnosis method that combines convolutional neural networks and image texture features. The main steps include: Firstly, ultrasound thyroid nodule dataset is established by collecting positive and negative samples, standardizing of images and segmentation of nodule area. Secondly, through texture features extraction, feature selection and data dimensionality reduction, texture features model is obtained; Thirdly, by transfer learning, deep neural network is used to obtain feature model of the nodule in images; Then, texture features model and convolutional neural network feature model are combined to form a new nodule feature model called Feature Fusion Network; Finally, Feature Fusion Network is applied to train and improve performance than single network, and a deep neural network diagnosis model that can adapt to the characteristics of thyroid nodules is built. In order to test this method, 1874 groups of clinical ultrasound thyroid nodules are collected. Harmonic average F-score based on Precision and Recall is used as an evaluation indicator. Experimental results show that Feature Fusion Network can distinguish between benign and malignant thyroid nodules with an F-score of 92.52%. Compared with traditional machine learning methods and convolutional neural networks, performance of this work is better.

Clinical procedures, which require a large number of personnel and medical resources, receive the majority of the current focus on thyroid nodule diagnosis. An automated thyroid ultrasound nodule identification system is built using image texture data and convolutional neural networks in this study. The following are the major phases: The underlying stages in building a ultrasound thyroid knob dataset incorporate gathering positive and negative examples, normalizing pictures, and portioning the knob region. Second, a texture features model is built by selecting features, reducing the dimensionality of the data, and extracting texture features. Third, deep neural networks in move learning are utilized to create an element model of the knob in an image. The convolutional brain network highlight model and the surface component model were combined to create the brand-new knob include model known as the Feature Fusion Network. The Feature Fusion Network is used to prepare and improve performance over a single organization in order to create a demonstrative model for deep neural networks that can adapt to a variety of knob features. 1874 clinical ultrasonography thyroid knobs were gathered for this investigation. The musical normal F-score considering Accuracy and Review is utilized as an assessment metric. With an F-score of 92.52 percent, the study’s findings suggest that the Element Combination Organization can differentiate between benign and harmful thyroid knobs. As far as execution, this methodology performs better compared to standard ML procedures and convolutional neural networks.

Marine oil spills are one of the major threats to marine ecological safety, and the rapid identification of oil films is of great significance to the emergency response. Marine radar can provide data for marine oil spill detection; however, to date, it has not been commonly reported. Traditional marine radar oil spill research is mostly based on grayscale segmentation, and its accuracy depends entirely on the selection of the threshold. With the development of algorithm technology, marine radar oil spill extraction has gradually come to focus on artificial intelligence, and the study of oil spills based on machine learning has begun to develop. Based on X-band marine radar images collected from the Dalian 716 incident, this study used image texture features, the BP neural network classifier, and threshold segmentation for oil spill extraction. Firstly, the original image was pre-processed, to eliminate co-channel interference noise. Secondly, texture features were extracted and analyzed by the gray-level co-occurrence matrix (GLCM) and principal component analysis (PCA); then, the BP neural work was used to obtain the effective wave region. Finally, threshold segmentation was performed, to extract the marine oil slicks. The constructed BP neural network could achieve 93.75% classification accuracy, with the oil film remaining intact and the segmentation range being small; the extraction results were almost free of false positive targets, and the actual area of the oil film was calculated to be 42,629.12 m2. The method proposed in this paper can provide a reference for real-time monitoring of oil spill incidents.

UAV imaging provides an efficient and non-destructive tool for characterizing farm information, but the quality of the UAV model is often affected by the image’s spatial resolution. In this paper, the predictability of models established using UAV multispectral images with different spatial resolutions for nitrogen content of winter wheat was evaluated during the critical growth stages of winter wheat over the period 2021–2022. Feature selection based on UAV image reflectance, vegetation indices, and texture was conducted using the competitive adaptive reweighted sampling, and the random forest machine learning method was used to construct the prediction model with the optimized features. Results showed that model performance increased with decreasing image spatial resolution with a R2, a RMSE, a MAE and a RPD of 0.84, 4.57 g m−2, 2.50 g m−2 and 2.34 (0.01 m spatial resolution image), 0.86, 4.15 g m−2, 2.82 g m−2 and 2.65 (0.02 m), and 0.92, 3.17 g m−2, 2.45 g m−2 and 2.86 (0.05 m), respectively. Further, the transferability of models differed when applied to images with coarser (upscaling) or finer (downscaling) resolutions. For upscaling, the model established with the 0.01 m images had a R2 of 0.84 and 0.89 when applied to images with 0.02 m and 0.05 m resolutions, respectively. For downscaling, the model established with the 0.05 m image features had a R2 of 0.86 and 0.83 when applied to images of 0.01 m and 0.02 m resolutions. Though the image spatial resolution affects image texture features more than the spectral features and the effects of image spatial resolution on model performance and transferability decrease with increasing plant wetness under irrigation treatment, it can be concluded that all the UAV images acquired in this study with different resolutions could achieve good predictions and transferability of the nitrogen content of winter wheat plants.

Hardness is one of the important attributes in determining the quality of dried fruits. Hardness assessment is normally carried out by manual inspection. This method is time consuming, laborious, expensive and subjective. The objective of this study was to develop a computer vision system with a monochrome camera to classify dates based on hardness. Date samples were obtained from three different growing regions in Oman and graded into soft, semi-hard, and hard classes based on hardness. A total of 1800 date samples were imaged individually using a monochrome camera (600 dates / class). Histogram and texture features were extracted from the acquired monochrome images and used in the classification models. The overall classification accuracies in three class model (soft, semi-hard, and hard) were 66% and 71% for linear discriminant analysis (LDA) and artificial neural network (ANN), respectively. It was improved to 84% and 77% in LDA and ANN, respectively while using two class model (soft and hard (semi-hard and hard together)). The histogram features were more contributing in the date classification based on hardness than image texture features. Computer vision technique has great potential to develop online quality monitoring systems for dates and other dried fruits.

Soil is an important environment for crop growth. Quick and accurately access to soil nutrient content information is a prerequisite for scientific fertilization. In this work, hyperspectral imaging (HSI) technology was applied for the classification of soil types and the measurement of soil total nitrogen (TN) content. A total of 183 soil samples collected from Shangyu City (People’s Republic of China), were scanned by a near-infrared hyperspectral imaging system with a wavelength range of 874–1734 nm. The soil samples belonged to three major soil types typical of this area, including paddy soil, red soil and seashore saline soil. The successive projections algorithm (SPA) method was utilized to select effective wavelengths from the full spectrum. Pattern texture features (energy, contrast, homogeneity and entropy) were extracted from the gray-scale images at the effective wavelengths. The support vector machines (SVM) and partial least squares regression (PLSR) methods were used to establish classification and prediction models, respectively. The results showed that by using the combined data sets of effective wavelengths and texture features for modelling an optimal correct classification rate of 91.8%. could be achieved. The soil samples were first classified, then the local models were established for soil TN according to soil types, which achieved better prediction results than the general models. The overall results indicated that hyperspectral imaging technology could be used for soil type classification and soil TN determination, and data fusion combining spectral and image texture information showed advantages for the classification of soil types.

Abstract Background The localization of proteins to specific subcellular structures in eukaryotic cells provides important information with respect to their function. Fluorescence microscopy approaches to determine localization distribution have proved to be an essential tool in the characterization of unknown proteins, and are now particularly pertinent as a result of the wide availability of fluorescently-tagged constructs and antibodies. However, there are currently very few image analysis options able to effectively discriminate proteins with apparently similar distributions in cells, despite this information being important for protein characterization. Findings We have developed a novel method for combining two existing image analysis approaches, which results in highly efficient and accurate discrimination of proteins with seemingly similar distributions. We have combined image texture-based analysis with quantitative co-localization coefficients, a method that has traditionally only been used to study the spatial overlap between two populations of molecules. Here we describe and present a novel application for quantitative co-localization, as applied to the study of Rab family small GTP binding proteins localizing to the endomembrane system of cultured cells. Conclusions We show how quantitative co-localization can be used alongside texture feature analysis, resulting in improved clustering of microscopy images. The use of co-localization as an additional clustering parameter is non-biased and highly applicable to high-throughput image data sets.

The robustness of image features is a very important consideration in quantitative image analysis. The objective of this paper is to investigate the robustness of a range of image texture features using hematoxylin stained breast tissue microarray slides which are assessed while simulating different imaging challenges including out of focus, changes in magnification and variations in illumination, noise, compression, distortion, and rotation. We employed five texture analysis methods and tested them while introducing all of the challenges listed above. The texture features that were evaluated include co-occurrence matrix, center-symmetric auto-correlation, texture feature coding method, local binary pattern, and texton. Due to the independence of each transformation and texture descriptor, a network structured combination was proposed and deployed on the Rutgers private cloud. The experiments utilized 20 randomly selected tissue microarray cores. All the combinations of the image transformations and deformations are calculated, and the whole feature extraction procedure was completed in 70 minutes using a cloud equipped with 20 nodes. Center-symmetric auto-correlation outperforms all the other four texture descriptors but also requires the longest computational time. It is roughly 10 times slower than local binary pattern and texton. From a speed perspective, both the local binary pattern and texton features provided excellent performance for classification and content-based image retrieval.