Similar Texture Research Articles

Overview of Essential Components in deep learning reference-based super resolution methods

Reference-based super resolution (RefSR) aims to recover the lost details in a low-resolution image and generate a high-resolution result, guided by a high-resolution reference image with similar contents or textures. In contrast to the traditional single-image super-resolution, which focuses on the intrinsic properties of the single low-resolution image, the challenge of RefSR lies in matching and aggregating highly-related but misaligned reference images with low-resolution images. Several effective but complex designs have been proposed to address this challenge, which poses difficulties in implementing RefSR in real-world applications. In order to better understand the working mechanism of RefSR and design a more efficient and lightweight architecture, we provide a review about the essential components of the existing deep learning-based RefSR. We decompose and classify the common pipeline into four submodules according to the functionalities. Then, we summarize and describe the implementation details of the commonly-adopted approaches in each submodule. Finally, we discuss the challenges and promising research directions.

Petrographic and Structural Study of Stalactites and Stalagmites of Mfamosing Camp II Cave in Cross River State, South Eastern Nigeria

This study investigated the petrographic and structural features of Mfamosing camp II cave stalactites and stalagmites. The cave is in the karst region of Udukpani in Cross River State, Southeast Nigeria. The objective of the study was to gain insights into the mineral composition, texture and structure of the stalactites and stalagmites of the cave through petrographic analysis. The study employed a methodology that involved field study, collection of representative samples, preparation of thin sections, and examination under a petrographic microscope. Field observations revealed that the stalactite and stalagmites were leucocratic in color. The rocks exhibited similar texture which was fine grained. The cave contained large number of stalactites and stalagmites of varrying lengths (icicle and cone dome shapes) with stromalitic features on the wall. The petrographic analysis revealed the predominance of carbonates (calcite) as the primary mineral. The average estimated minerals composition present were 54.5% of Calcite, 5.3% of Quartz, and 2% of feldspar for stalactites. 51% of Calcite, 5% of Quartz, and 2.67% of feldspar for stalagmites. The results shows that the major mineral composition of the speleothems the cave are CaO.

Dough rheology and internal structure of bread produced with wheat flour partially substituted by buckwheat flour: A step towards enhancing nutritional value

Buckwheat flour is a pseudocereal that can be used to improve the nutritional value of staple food, such as bread. However, substituting wheat flour can be challenging due to potential loss of the final product's technological quality. Therefore, understanding the influence of buckwheat inclusion level on dough rheology and bread quality is essential. This study evaluated the effect of wheat flour substituted by buckwheat flour on dough rheology and bread quality over the staling process. Wheat flour was replaced by 0%, 10%, 20% and 30% of buckwheat flour and the resulting dough mixing properties, viscosity profile, gel texture, rheology behaviour and extensibility properties were analysed. Internal bread structure (C-Cell and confocal microscopy), specific volume, density and crumb texture profile were also assessed. The increase in buckwheat level gradually decreased flour water absorption (p < 0.05), dough stability and development time, with increased gluten weakening as evidenced by decreased C2 values (protein strength) of Mixolab. As a result, the dough was less resistant to extension, with decreased extensibility. The highest viscosity profile of the starch paste was observed with 30% buckwheat inclusion level, which may lead to faster bread staling process. A solid-like and elastic behaviour was observed in the rheology evaluation, suggesting an interaction between fibre and starch from buckwheat with gluten proteins. Although changes in dough rheology occurred, the bread quality remained unchanged at levels of up to 20% buckwheat level (p > 0.05), with similar specific volume, density, internal structure and texture profile compared to the control bread (0% buckwheat). In conclusion, buckwheat was found to be a viable wheat flour substitute with only minimal changes in dough behaviour and maintenance of bread technological quality, while enhancing the nutritional profile by increasing fibre content and total essential amino acid.

“DAUNTLESS” BUSANA DENGAN SUMBER IDE PAVILION CENGHUANG

The process of creating this artwork is explored as a cultural topic, and ideas are sourced from the Chenghuang Pavilion. The inspiration for this artwork comes from the pavilion, which has a strong Chinese cultural identity. The iconic form of the pavilion and the bold colors present in it add a sense of power, representing a strong and characteristic pavilion. The process of creating this artwork involves several analyses, starting from the mood board, design, calculation, selection of materials and techniques, and design simulation. The process of creating the artwork begins with the search for ideas and concepts from a similar pavilion, which has a similar architectural style and texture to the Chenghuang Pavilion. Until a mature concept is found, the selection of ideas, design, techniques, and creation of the pavilion clothing is carried out. Thus, a suitable artwork is obtained that aligns with the purpose, reasoning, and analysis of why the product was created.

Analisis Kualitas Abon Berbahan Dasar Carica

Abon is a type of processed food that has a long storage time. Abon meat is usually processed from shredded animal meat which is then seasoned with spices and fried and then drained until it is completely dry. Abon can be created by changing the raw materials into other raw materials with similar textures. Other raw materials can be in the form of vegetables or fruits depending on needs. Abon made from carica fruit is a new innovation to increase the nutritional quality of shredded and the selling value of carica fruit, which is often found in the Wonosobo area. This research is an experimental type that aims to find out how the quality of carica-based shredded meat is seen from the aspect of aroma, taste, color, texture and appearance which was tested on 2 experts and 30 panelists chosen by researchers with various considerations. The analytical method used by the researcher is descriptive statistics which is done by tabulating the results of the questionnaire in the form of a single table. The table in question can provide a general description of the panelists' responses to the quality of shredded carica based on the tendency of the rating results, then interpretation is carried out to understand the meaning behind this tendency. Based on the results of the research and the results of panelist interviews on the quality of abon carica on the aspects of aroma, taste, color, and texture, the average value of 5 or which means "strongly agree". This is also in line with the results of interviews with all selected experts who stated that they liked and agreed with all aspects. It can be concluded that shredded carica has good quality based on aspects of aroma, taste, color, texture, and appearance.

Segmentation and tracking of vegetable plants by exploiting vegetable shape feature for precision spray of agricultural robots

AbstractFor robotic precision spray application in vegetable farms, simultaneous accurate instance segmentation and robust tracking of plants are of great importance and a prerequisite for the following spray action. With onboard cameras, agricultural robots can apply Multiple Object Tracking and Segmentation (MOTS) methods, for instance, segmentation and tracking of plants. By assigning a unique identification for each vegetable, it ensures the robot to spray each vegetable exactly once, while traversing along the farm rows. Conventional MOTS methods, which are mostly designed for tracking pedestrians or vehicles, usually extract their color and texture features for associating different targets in consecutive images. However, vegetable plants of the same species normally show similar color and texture, which leads to degraded performance when conventional MOTS methods are used. To solve the challenging problem of associating vegetables with similar color and texture in consecutive images, in this paper, a novel MOTS method that exploits contour and blob features is proposed, for instance, segmentation and tracking of multiple vegetable plants. The method takes advantage of the fact that different plants normally possess different shape contours and blob properties. With images captured on top of them, these features of the same plant show little difference in consecutively captured images. Comprehensive experiments including ablation studies are conducted, which prove its superior performance over two state‐of‐the‐art MOTS methods. Compared with the conventional MOTS methods, the proposed method is able to re‐identify objects which have gone out of the camera field of view and re‐appear again using the proposed data association strategy, which is important to ensure each vegetable be sprayed only once when the robot travels back and forth. Although the method is tested on lettuce farm, it can be applied to other similar vegetables, such as broccoli and canola. Both the code and the dataset of this paper are publicly released for the benefit of the community: https://github.com/NanH5837/LettuceMOTS.

Research on She nationality clothing recognition based on color feature fusion with PSO-SVM

Abstract Although the color characteristics of She nationality clothing are slightly different, there are multiple similarities in shapes and textures. Therefore, it is difficult to effectively distinguish different branches of She nationality clothing. To address this problem, this article, taking into account color feature fusion, proposes a recognition method based on a hybrid algorithm of particle swarm optimization and support vector machine (PSO-SVM). First, the color histogram and color moment (CM) feature descriptors were extracted from the five branches of She nationality clothing, and the color feature distribution of each branch was obtained. Then, color feature fusion is performed through optimization and dimensionality reduction of principal components. Furthermore, PSO was introduced to independently optimize parameter combinations. Finally, the different branches of She nationality clothing were automatically recognized. The results demonstrated that the proposed method could effectively distinguish different branches of She nationality clothing. Compared with the recognition accuracy of approaches using single-color histogram and CM feature, the performance of our proposed method was increased by 5.25 and 6.44%, respectively. When the penalty parameter γ \gamma and kernel parameter δ 2 {\delta }^{2} of SVM were 123.29 and 1.16, respectively, the recognition accuracy of the model was the highest, reaching 98.67%. The proposed method could be a reference for the subdivision recognition of She nationality clothing.

Kadar Serat dan Tekstur Kue Kembang Goyang dengan Substitusi Beras Merah Organik

Kembang goyang is Indonesian traditional snack, but it has quite low of dietary fiber and resistant starch content. Organic and nonorganic red rice could be used to increase level of dietary fiber by transforming it into red rice flour and used as a substitute of white rice flour in the making of kembang goyang snack. The purpose of this study is to determine type and substitution ratio of red rice flour into white rice flour to produce the best formulation of kembang goyang snack based on its dietary fiber content and texture. Kembang goyang snack is made by organic and nonorganic red rice flour with different ratios of red rice flour: white rice flour (0:100; 20:80; 40:60; 60:40; and 80:20). The result showed that organic red rice flour has lower yield (91,47 ± 0,23%) compared to nonorganic red rice flour's yield (92,80 ± 0,09%). In contrast, the dietary fiber content (8,09 ± 0,04%) and resistant starch content (3,95 ± 0,00%) of organic red rice flour is higher than the dietary fiber content (6,46 ± 0,07%) and resistant starch content (1,76 ± 0,03%) of non-organic red rice flour. Overall, the moisture content for both types of red rice flour ranged between 5,91 ± 0,08 to 6,35 ± 0,60%. The best formulation of kembang goyang snack is obtained by using organic red rice flour at ratio 40:60 of red rice flour and white rice flour. The product is classified as fiber-rich product and has similar texture (toughness and fracturability) to commercial product with 6,30 ± 0,01% of resistant starch content.

Remote Sensing Image Super-Resolution via Multi-Scale Texture Transfer Network

As the degradation factors of remote sensing images become increasingly complex, it becomes challenging to infer the high-frequency details of remote sensing images compared to ordinary digital photographs. For super-resolution (SR) tasks, existing deep learning-based single remote sensing image SR methods tend to rely on texture information, leading to various limitations. To fill this gap, we propose a remote sensing image SR algorithm based on a multi-scale texture transfer network (MTTN). The proposed MTTN enhances the texture feature information of reconstructed images by adaptively transferring texture information according to the texture similarity of the reference image. The proposed method adopts a multi-scale texture-matching strategy, which promotes the transmission of multi-scale texture information of remote sensing images and obtains finer-texture information from more relevant semantic modules. Experimental results show that the proposed method outperforms state-of-the-art SR techniques on the Kaggle open-source remote sensing dataset from both quantitative and qualitative perspectives.

Early dextral shear motion during the breakup of East Gondwana documented in clastic dykes in the Tethys Himalaya

Much of the geological record of the northern Indian passive continental margin has been lost due to erosion or obscured due to tectonic burial associated with the India-Eurasia collision. Consequently, tectonic models of India's breakup from East Gondwana incorporate limited data from the northern margin of Greater India. We present evidence of the Late Jurassic clastic dykes crosscutting contemporaneous sedimentary sequences in the Tethys Himalaya. We propose that the original orientation of the clastic dykes provides kinematic information about Greater India's breakup from East Gondwana. The clastic dykes are meters to hundreds of meters long and centimeters to meters wide. They taper downwards and crosscut country rocks at high angles. The clastic dykes and their host strata are quartzose sandstones and share similar textures and detrital zircon age populations (dominantly 600–500 Ma and 1000–800 Ma zircon grains). These similarities indicate the clastic dykes and their host rocks share a common source as well as comparable transport and deposition dynamics. Notably, the clastic dykes lack distinct signs of ptygmatic folds, shale/mudrock clasts, surface orientated structures, or bifurcations on dyke margins. Decompaction and structural restoration of the sequences show that the sandstone dykes/fractures and en échelon dyke arrays can be classified into three distinct orientation groups NW (349° − 295°), SW (∼210°) and NE (∼40°). There is no apparent difference in the age or provenance of the clastic material within the different orientation groups. We propose that the clastic dykes signify passively infilled tensile fractures formed within a NW-SE dextral strike-slip megashear system during the Late Jurassic. This system likely developed in a transform zone within Greater India prior to the separation of the Indian Plate from East Gondwana.

Multi-Domain Rapid Enhancement Networks for Underwater Images

Images captured during marine engineering operations suffer from color distortion and low contrast. Underwater image enhancement helps to alleviate these problems. Many deep learning models can infer multi-source data, where images with different perspectives exist from multiple sources. To this end, we propose a multichannel deep convolutional neural network (MDCNN) linked to a VGG that can target multi-source (multi-domain) underwater image enhancement. The designed MDCNN feeds data from different domains into separate channels and implements parameters by linking VGGs, which improves the domain adaptation of the model. In addition, to optimize performance, multi-domain image perception loss functions, multilabel soft edge loss for specific image enhancement tasks, pixel-level loss, and external monitoring loss for edge sharpness preprocessing are proposed. These loss functions are set to effectively enhance the structural and textural similarity of underwater images. A series of qualitative and quantitative experiments demonstrate that our model is superior to the state-of-the-art Shallow UWnet in terms of UIQM, and the performance evaluation conducted on different datasets increased by 0.11 on average.

Thermally stable piezoelectric sensors for quantitative pressure sensing based on linear piezoelectric zinc oxide thin films

Linear piezoelectric materials offer new opportunities for next-generation multifunctional micro/nano systems on chip aside from ferroelectrics. However, the advantage of quantitative pressure sensing of linear piezoelectric materials over ferroelectrics has not been demonstrated, while the thermal stability of linear piezoelectric zinc oxide (ZnO) is yet to be evaluated. In this work, pressure sensors based on linear piezoelectric undoped ZnO and ferroelectric vanadium doped ZnO (V-ZnO) films are studied side by side. Both films possess similar fiber textures oriented along the polar ZnO [002] direction, yet each of the V-ZnO grains consists of strip-like domains observed by SEM. Applying triangular-wave pressures, the linear piezoelectric ZnO sensor generates triangular-wave voltage signals that are one-to-one linearly correlated to the applied pressure, showing decisive advantages for quantitative pressure sensing compared to the ‘sail-like’ voltage signals of the ferroelectric V-ZnO sensor that is many-to-one correlated to the applied pressure. Superior thermal stability is demonstrated in the linear piezoelectric ZnO sensor showing stable pressure sensing properties up to 450 °C compared to the degraded sensing properties of the ferroelectric V-ZnO sensor at 150 °C. The thermally stable piezoelectric sensors for quantitative pressure sensing based on linear piezoelectric ZnO films serve as candidates for high-performance multifunctional micro/nano systems.

Automatic myeloblast segmentation in acute myeloid leukemia images based on adversarial feature learning

The diagnosis of acute myeloid leukemia (AML) requires reliable detection and counting of immature white blood cells called myeloblasts on bone marrow aspirate and biopsy slides. Experienced hematologists/oncologists typically spend many hours manually reviewing blood cells and performing a microscopic morphological examination of myeloblasts on glass slides to determine this diagnosis. While manual counting of myeloblasts is considered the gold standard for an accurate diagnosis of AML, the process is time-consuming and difficult to standardize due to high intra- and inter-observer variability. Machine learning algorithms coupled with deep learning techniques represent potentially promising approaches for automated counting. Since myeloblasts possess very similar texture patterns to other hematopoietic cells (OHCs), accurately differentiating myeloblasts from other blood cells is a challenge for most segmentation models. We present a new conditional generative adversarial network (cGAN)-based segmentation model (AMLcGAN) to efficiently segment myoblasts from slides of patients with AML. Validation of the approach on 204 cytological images reveals a mean dice coefficient (mDICE) of 82.51 % with respect to manual segmentations. We also compared the new AMLcGAN method with three deep learning models: Unet, Segnet and Pix2pix, the corresponding mDICE values were 77.4 %, 78.2 %, and 79.53 %, respectively. We also employed the segmentation results from 4 different models to extract blast statistics feature, Haralick texture feature, Fractal dimension feature and myeloblast shape features to create 4 different prognostic models to predict risk of recurrence in AML patients. TheConcordance Index values for AMLcGAN and the three other models respectively were 0.63, 0.53, 0.58, and 0.56. The comprehensive evaluation results presented suggest that the AMLcGAN approach can enable accurate blast segmentation and for construction of accurate prognostic models for predicting risk of recurrence in AML patients.

Application of bilateral facial perforator artery flap in repairing large area defect in middle and lower part of nose

To investigate the feasibility and effectiveness of bilateral facial perforator artery flap in repairing large area defect in middle and lower part of nose. The clinical data of 18 patients with large area defect in middle and lower part of nose repaired by bilateral facial perforator artery flap between January 2019 and December 2022 were retrospectively analyzed. Among them, there were 13 males and 5 females, the age ranged from 43 to 81 years, with an average of 63 years. There were 3 cases of nasal trauma, 4 cases of basal cell carcinoma, 8 cases of squamous cell carcinoma, 1 case of lymphoma, and 2 cases of large area solar keratosis. The size of the defect ranged from 3.0 cm×3.0 cm to 4.5 cm×4.0 cm; the size of unilateral flap ranged from 3.0 cm×1.3 cm to 3.5 cm×2.0 cm, and the size of bilateral flaps ranged from 3.3 cm×2.6 cm to 4.5 cm×4.0 cm. One patient developed skin flap necrosis after operation, and a frontal skin flap was used to repair the wound; 1 case gradually improved after removing some sutures due to venous congestion in the skin flap, and the wound healing was delayed after dressing change; the remaining 16 cases of bilateral facial perforator artery flaps survived well and all wounds healed by first intention, without any "cat ear" malformation. All 18 patients had first intention healing in the donor area, leaving linear scars without obvious scar hyperplasia, and no facial organ displacement. All patients were followed up 3-12 months, with an average of 6 months. Due to the appropriate thickness of the flap, none of the 18 patients underwent secondary flap thinning surgery. All flaps had good blood circulation, similar texture and color to surrounding tissues, symmetrical bilateral nasolabial sulcus, and high patient satisfaction. The bilateral facial perforator artery flaps for repairing large area defect in middle and lower part of nose can achieve good appearance and function, and the operation is relatively simple, with high patient satisfaction.

Achiral Dichroic Dyes-mediated Circularly Polarized Emission Regulated by Orientational Order Parameter through Cholesteric Liquid Crystals.

It is noteworthy that cholesteric liquid crystal (CLC) platforms have been witnessed in high-performance circularly polarized luminescence (CPL) behaviors through the highly organized chiral co-assembled arrangement of achiral dyes. However, most CPL-active design strategies are closely relative to the helix co-assembly structure of CLC rather than achiral dyes. Herein, we developed an intriguing regulation strategy for CPL-active CLC materials. They were regulated using the orientational order parameter (SF) of achiral dichroic dyes as an incisive probe for the order arrangement degree of achiral dyes in CLC media. The I-shaped phenothiazine derivative PHECN dye (SF =0.30) emitted a strong CPL signal (|glum| = 0.47). In contrast, the T-shaped derivative (PHEBen) dye (SF =0.09) showed a weak circular polarization level (|glum| = 0.07) at similar CLC textures. Most interestingly, this kind of dichroic PHECN dye with a higher SF could greatly improve the contrast ratio of CPL (Δglum = 0.47) and emission intensity (ΔFL = 46.0%) at direct-current electric field compared with the T-shaped PHEBen (Δglum = 0.07 and ΔFL = 1.0%) in CLCs. This work demonstrates that an induced CPL emission can be mediated using achiral dichroic dye, which will open a new avenue for developing excellent CPL-active display materials.

Achiral Dichroic Dyes‐mediated Circularly Polarized Emission Regulated by Orientational Order Parameter through Cholesteric Liquid Crystals

AbstractIt is noteworthy that cholesteric liquid crystal (CLC) platforms have been witnessed in high‐performance circularly polarized luminescence (CPL) behaviors through the highly organized chiral co‐assembled arrangement of achiral dyes. However, most CPL‐active design strategies are closely relative to the helix co‐assembly structure of CLC rather than achiral dyes. Herein, we developed an intriguing regulation strategy for CPL‐active CLC materials. They were regulated using the orientational order parameter (SF) of achiral dichroic dyes as an incisive probe for the order arrangement degree of achiral dyes in CLC media. The I‐shaped phenothiazine derivative PHECN dye (SF=0.30) emitted a strong CPL signal (|glum|=0.47). In contrast, the T‐shaped derivative (PHEBen) dye (SF=0.09) showed a weak circular polarization level (|glum|=0.07) at similar CLC textures. Most interestingly, this kind of dichroic PHECN dye with a higher SF could greatly improve the contrast ratio of CPL (Δglum=0.47) and emission intensity (ΔFL=46.0 %) at direct‐current electric field compared with the T‐shaped PHEBen (Δglum=0.07 and ΔFL=1.0 %) in CLC. This work demonstrates that an induced CPL emission can be mediated using achiral dichroic dye, which will open a new avenue for developing excellent CPL‐active display materials.

Potential of hydrolyzed wheat protein in soy-based meat analogues: Rheological, textural and functional properties

Hydrolyzed proteins, which are considered to possess significant bioactive properties such as antioxidant and high digestibility, have garnered increasing interest as food ingredients. This study investigates the feasibility of using hydrolyzed wheat gluten (HWG) and soy protein concentrate (SPC) in various ratios to create meat analogues using high-moisture extrusion technology. Results indicate that meat analogues with 40% HWG addition to SPC have a better texture and greater similarities in terms of hardness, chewiness, and toughness to chicken meat than meat analogues with 40% wheat gluten (WG) addition to SPC. Additionally, the meat analogues with HWG showed high antioxidant capacity, protein digestibility, and amino acid composition, indicating potential health benefits. These findings indicate that HWG could serve as a texture modifier to improve both the texture and nutritional content of meat analogues.

Discontinuous yielding behavior, strengthening effect and deformation mechanism of double wall copper-brazed steel tube under various continuous sectional cooling

Unlike traditional cooling, continuous sectional cooling (CSC) consists of various cooling modes, which is of decisive significance to the microstructure and performance of double wall copper-brazed steel tube (DWBT). In this work, three CSC (CSC-A/B/C) processes are chosen to reveal strengthening and deformation mechanisms of DWBT. The results show that there are significant differences in discontinuous yield behavior, natural aging phenomenon, and tensile properties of the DWBT among three CSC processes due to diverse grain size and precipitates, but a similar texture is obtained due to Kurdjumov-Sachs relationship. For discontinuous yielding, the propagation velocity of Lüders band is a key factor to determine the yield point elongation (YPE), which reduces with decreasing grain sizes due to the interaction between dislocation and grain boundary. Thus, the CSC-C (1120 °C→4s940 °C…360 °C→16s110 °C) with small grains has a longer YPE. Meanwhile, the precipitation strengthening dominates the highest strength and lowest elongation of the CSC-A (1120 °C…100 °C) due to abundant MC carbides. Compared with CSC-B (1120 °C…170 °C), the yield strength of CSC-C is higher while ultimate tensile strength is lower, which is attributed to different contributions of precipitation and grain boundary strengthening to both. The MC carbide causes the natural aging of DWBT. Moreover, the geometrically necessary dislocations (GNDs) and low angle grain boundaries (LAGBs) raise linearly with strain. The GNDs prefer to be distributed in small grains, and uniform grain can weaken the difference of GNDs distribution in different grain sizes. The LAGBs including dense dislocation walls or microbands are formed by the grain subdivision mechanism.

Unsupervised Deep Learning Approach for Characterizing Fractality in Dried Drop Patterns of Differently Mixed Viscum album Preparations

This paper presents a novel unsupervised deep learning methodology for the analysis of self-assembled structures formed in evaporating droplets. The proposed approach focuses on clustering these structures based on their texture similarity to characterize three different mixing procedures (turbulent, laminar, and diffusion-based) applied to produce Viscum album Quercus 10−3 according to the European Pharmacopoeia guidelines for the production of homeopathic remedies. Texture clustering departs from obtaining a comprehensive texture representation of the full texture patch database using a convolutional neural network. This representation is then dimensionally reduced to facilitate clustering through advanced machine learning techniques. Following this methodology, 13 clusters were found and their degree of fractality determined by means of Local Connected Fractal Dimension histograms, which allowed for characterization of the different production modalities. As a consequence, each image was represented as a vector in R13, enabling classification of mixing procedures via support vectors. As a main result, our study highlights the clear differences between turbulent and laminar mixing procedures based on their fractal characteristics, while also revealing the nuanced nature of the diffusion process, which incorporates aspects from both mixing types. Furthermore, our unsupervised clustering approach offers a scalable and automated solution for analyzing the databases of evaporated droplets.

Machine-learning-assisted classification of construction and demolition waste fragments using computer vision: Convolution versus extraction of selected features

Improper sorting of construction and demolition waste (CDW) leads to significant environmental and economic implications, including inefficient resource use and missed recycling opportunities. To address this, we developed a machine-learning-assisted procedure for recognizing CDW fragments using an RGB camera. Our approach uniquely leverages selected feature extraction, enhancing classification speed and accuracy. We employed three classifiers: convolutional neural network (CNN), gradient boosting (GB) decision trees, and multi-layer perception (MLP). Notably, our method’s extraction of selected features for GB and MLP outperformed the traditional CNN in terms of speed and accuracy, especially for challenging samples with similar textures. Specifically, while convolution resulted in an overall accuracy of 85.9%, our innovative feature extraction approach yielded accuracies up to 92.3%. This study’s findings have significant implications for the future of CDW management, offering a pathway for efficient and accurate waste sorting, fostering sustainable resource use, and reducing the environmental impact of CDW disposal. Supplementary materials, including datasets, codes, and models, are provided, promoting transparency and reproducibility.