Human Interference Research Articles

Animal Detection based Smart Farming in Animal Repellent Using AI and Deep Learning

The combination of artificial intelligence (AI) and deep learning in agriculture has ushered in a new era in agriculture with new solutions designed to solve many problems. This paper presents an animal killing system for smart agriculture that uses artificial intelligence and deep learning to reduce the growing problem of animal damage. As the world population continues to grow, increasing food availability is important; Therefore, it is critical to protect crops from wild animals and pests. Deforestation due to livestock farming has become one of the largest human-wildlife conflicts due to human interference with habitats and deforestation. Wild animals can kill farmers working in the fields, causing major crop losses. Farmers suffered huge crop losses due to wild animals such as elephants, wild boars and deer attacking agriculture. Protecting crops from wild animals is one of the biggest concerns of today's farmers. There are many ways to solve this problem, both lethal (such as shooting and trapping) and non-lethal. (such as railings, pesticides, organic matter, netting or electric fencing). The sensor rotates around the lens and uses DCNN software to detect the target. If an animal is found, it sends a message to the Animal Repellent Module with information on the type of ultrasound that should be produced based on the animal. The development of drones with controlled flight control, as well as lightweight and powerful hyperspectral snapshot cameras that can be used to calculate crop biomass growth and fertilization status, responds to complex agricultural management strategies. KEYWORDS: Animal detection, VGG-Net, Bi-LSTM, convolutional neural network, activity recognition, video surveillance, wild animal monitoring, alert system.

The Effects of Wetland Degradation on Ecological Species

Wetlands are vital ecosystems that provide numerous ecological services, including flood control, water filtration, carbon sequestration, and habitat for diverse flora and fauna. However, human activities such as urbanization, agriculture, and infrastructure development have led to widespread wetland degradation worldwide. Wetland ecosystems diminished by 21-35% between 1700 and 2020 as a result of human interference, with at least 1.3 million square miles of wetlands lost globally—an area about the size of Alaska, Texas, California, Montana, New Mexico, and Arizona combined. This research used a documented methodology for extracting information from different books, conversations, conferences, and international organizations to understand the effects of wetland degradation on the decline of species and strategies for wetland conservation and restoration. This paper aims to elucidate the multifaceted effects of wetland degradation on ecological species. The results indicate that wetland degradation declines water quality, leading to alterations in water flow patterns, decreased groundwater recharge, and increased flooding downstream. This can result in the loss of biodiversity, as many species depend on specific water levels and habitats within wetlands. Moreover, the loss of wetlands diminishes their volume to store carbon, contributing to greenhouse gas emissions and exacerbating climate change. Furthermore, wetland degradation compromises water quality by reducing the natural filtration and purification functions of wetlands. Contaminants from agricultural runoff, industrial discharge, and urban pollution accumulate in degraded wetlands, posing risks to human health and aquatic ecosystems. Additionally, the loss of wetlands exacerbates erosion and sedimentation, leading to habitat destruction and loss of coastal resilience against storms and sealevel rise. Addressing wetland degradation requires a multifaceted approach, including policy interventions, restoration efforts, and public awareness campaigns. Effective wetland conservation strategies involve the preservation of existing wetlands, restoration of degraded ones, and sustainable management practices to mitigate further degradation. Collaborative efforts between governments, NGOs, local communities, and stakeholders are essential to safeguarding these critical ecosystems and the invaluable services they provide to the environment and society.

Identifying impactful sea turtle conservation strategies: a mismatch between most influential and most readily manageable life-stages

Sea turtles worldwide face a range of threats including sea level rise and warming associated with climate change, predation by invasive species, plastic and light pollution, coastal development, and human interference. Conservation managers have a long history of aiding sea turtle populations, from protecting nests to head-starting hatchlings. Due to these challenges faced by turtles, there is a constant focus on assessing the likely success of proposed conservation interventions to help inform decision-making processes. We develop an age-based, spatially implicit population model for the north-west shelf stock of Australia’s endemic flatback turtle Natator depressus that estimates the long-term outcomes of a range of onshore and offshore conservation interventions. Analysis of the model shows that young adults contribute most to population growth (i.e. have highest expected future reproductive success); however, this is often the most difficult life stage to manipulate in the field. Observable outcomes of interventions are often delayed for many years, especially for on shore activities (e.g. protecting eggs and hatchlings), due to late age to maturity. The potential impact of warming-induced female bias on population dynamics was also investigated. Although such bias increases population growth rates in the short term, negative effects of the bias (e.g. reduced female mating success) and negative environmental effects (e.g. reduced survival rates, habitat loss) can lead to sustained declines. Population models can rapidly assess climate change and conservation impacts on turtle dynamics and can guide monitoring efforts for real-world application.

The robot saw it coming: physical human interference, deservingness, and self-efficacy in service robot failures

ABSTRACT Robotic services’ popularity continues to increase due to technological advancements, labour shortages, and global crises. Yet, while providing these services, robots are subject to occasional physical interruption by humans to them, thus restricting their functioning and, at times, leading to failure. To investigate this issue, the present study examined the role of third-party human interference in service robot failures and its effects on the observers’ attitudes towards and willingness to engage with the robot. We manipulated human interference resulting in different robotic service failures in two online scenario-based experiments. The results revealed that individuals held less favourable attitudes towards a failed service robot without (vs. with) physical human interference, and they were less willing to engage with the failed service robot without (vs. with) physical human interference. The perceived deservingness of the robot accounted for this effect, moderated by the person’s self-efficacy regarding robots. The results are discussed with their implications for not only the theory of service failures and human-service robot interactions but also for robotic service providers.

An Interactive Feeder to Induce and Assess Emotions from Vocalisations of Chickens.

Understanding the emotional states of animals is a long-standing research endeavour that has clear applications in animal welfare. Vocalisations are emerging as a promising way to assess both positive and negative emotional states. However, the vocal expression of emotions in birds is a relatively unexplored research area. The goal of this study was to develop an interactive feeding system that would elicit positive and negative emotional states, and collect recordings of the vocal expression of these emotions without human interference. In this paper, the mechatronic design and development of the feeder is described. Design choices were motivated by the desire for the hens to voluntarily interact with the feeder and experience the different stimuli that were designed to induce (1) positive low-arousal, (2) positive high-arousal, (3) negative low-arousal, and (4) negative high-arousal states. The results showed that hens were motivated to engage with the feeder despite the risk of receiving negative stimuli and that this motivation was sustained for at least 1 week. The potential of using the interactive feeder to analyse chicken vocalisations related to emotional valence and arousal is being explored, offering a novel proof of concept in animal welfare research. Preliminary findings suggest that hens vocalised in response to all four stimulus types, with the number of vocalisations, but not the probability of vocalising, distinguishing between low- and high-arousal states. Thus, the proposed animal-computer interaction design has potential to be used as an enrichment device and for future experiments on vocal emotions in birds.

IOT Based Smart Crib

The Internet of Things, or IoT, gets its name because each device can send and receive data from one another. So it is similar to the actual Internet. However, you need to send and receive data on the Internet. And it only connects two computers. But in IoT, the network is between various devices or things. And they send and receive data without human interference. The number of IoT devices is increasing every day. In 2021, there were 10 billion IoT devices. Currently, the world has over 14.4 billion IoT devices. And after deploying 5G worldwide, this number may rise to over 75 billion

HomeOSD: Appliance Operating-Status Detection Using mmWave Radar.

Within the context of a smart home, detecting the operating status of appliances in the environment plays a pivotal role, estimating power consumption, issuing overuse reminders, and identifying faults. The traditional contact-based approaches require equipment updates such as incorporating smart sockets or high-precision electric meters. Non-constant approaches involve the use of technologies like laser and Ultra-Wideband (UWB) radar. The former can only monitor one appliance at a time, and the latter is unable to detect appliances with extremely tiny vibrations and tends to be susceptible to interference from human activities. To address these challenges, we introduce HomeOSD, an advanced appliance status-detection system that uses mmWave radar. This innovative solution simultaneously tracks multiple appliances without human activity interference by measuring their extremely tiny vibrations. To reduce interference from other moving objects, like people, we introduce a Vibration-Intensity Metric based on periodic signal characteristics. We present the Adaptive Weighted Minimum Distance Classifier (AWMDC) to counteract appliance vibration fluctuations. Finally, we develop a system using a common mmWave radar and carry out real-world experiments to evaluate HomeOSD's performance. The detection accuracy is 95.58%, and the promising results demonstrate the feasibility and reliability of our proposed system.

Differential responses of ecosystem stability to climatic and anthropogenic factors in connected and isolated lake basins on the Yangtze River

Maintaining optimal ecological security in the Yangtze River-connected and isolated lake basins is of great significance to national projects involving Yangtze River protection. Ecosystem stability and associated factors are important components of ecological security in these basins. However, few studies have focused on ecosystem stability and its driving factors over long periods in the Yangtze River Basin. In this study, a remote sensing index was used to analyze the spatiotemporal variation in the ecosystem stability of the Dongting Lake Basin (DTL), Poyang Lake Basin (PYL), and the isolated Chaohu Lake Basin (CHL) and Taihu Lake Basin (THL) in the Yangtze River over the period 2000–2022 to determine the potential affecting factors. The results showed fluctuations in the ecosystem stability of the DTL and PYL, while a V-shape was observed for the CHL and THL during the same period; the closer to the lake, the weaker the stability of the ecosystem, especially in the DTL and PYL. Moreover, the ecosystem stability was greater in the DTL and PYL than in the CHL and THL. The spillover effect of anthropogenic activities on the ecosystem stability of the four basins and the direct effect of temperature have the greatest effect on the ecosystem stability. Specifically, the ecosystem stability index for the area around the DTL and PYL decreased with increasing human interference, whereas the opposite was observed in the CHL and THL. The effect of temperature was negative for the ecosystem stability of DTL and PYL and significantly positive for CHL and THL, at a level of 0.01 %. The findings of this study provide significant information for targeted ecological restoration of the Yangtze River Basin.

Rain use efficiency gradients across Australian ecosystems

Rain use efficiency (RUE) quantifies the ecosystem's capacity to use precipitation water to assimilate atmospheric CO2. The spatial distribution of RUE and its drivers across the Australian continent is largely unknown. This knowledge gap limits our understanding of the possible contribution of Australian ecosystems to global carbon assimilation. This study investigates the spatial distribution of RUE across diverse terrestrial ecosystems in Australia. The results show that RUE ranges from 0.43 (1st percentile) to 3.10 (99th percentile) g C m−2 mm−1 with a continental mean of 1.19 g C m−2 mm−1. About 68 % of the spatiotemporal variability of RUE can be explained by a multiple linear regression model primarily contributed by climatic predictors. Benchmarked by the model estimation, drainage-diverging/converging landscapes tend to have reduced/increased RUE. The model also revealed the impact of increasing atmospheric CO2 concentration on RUE. The continental mean RUE would increase by between 29.3 and 64.8 % by the end of this century under the SSP5–8.5 scenario in which the CO2 concentration is projected to double from the present level. This increase in projected RUE is attributed to the assumed greening effect of increasing CO2 concentration, which does not consider the saturation of CO2 fertilisation effect and the warming effect on increasing wildfire occurrence. Under the SSP1–2.6 scenario, RUE would decrease by about 7 %. This study provides baseline RUEs of various ecosystems in Australia for investigating the impacts of human interferences and climate change on the capacity of Australian vegetation to assimilate atmospheric CO2 under given precipitation.

Assessing the Spatiotemporal Variations in the Hydrological Response of the Qin River Basin in Loess Plateau, China

In recent decades, the hydrological processes on the Loess Plateau have undergone significant changes due to the combined effects of climate change and human activities. This study investigates the Qin River basin, the core area of the Loess Plateau. Between 1971 and 2000, the hydrometeorology of the flood season in the Qin basin on the Loess Plateau in China changed abruptly in 1982 and 1992. This study uses distributed hydrologic modeling and statistical analysis to distinguish the spatial and temporal impacts of climate change and human activities. The fastest dynamic changes in land use occurred in grassland and construction land, accounting for 0.36% and −0.1%, respectively, from 1982 to 1991. From 1992 to 2000, it was 0.29% and 0.3%, respectively. The increase in basin area where the Normalized Difference Vegetation Index increased was 95%. However, except for significant changes in runoff from 1981 to 1992, there were no significant changes in precipitation, runoff, and temperature during other flood seasons. During the two studied variation periods, from 1982 to 1991 and from 1992 to 2000, the contribution rate of climate change to flood season runoff (CClimate) decreased from 83.26% to 74.47%, while the contribution rate of human activity disturbance to flood season runoff (Chuman) increased from 16.74% to 25.53%. In terms of spatial distribution, from 1982 to 1991, areas with a Chuman > 50% were primarily concentrated in the upper reaches of the basin, while from 1992–2000, these areas shifted to the lower reaches. This study offers valuable insights for water resource planning and soil conservation measures in the Loess Plateau and similar arid regions worldwide, aiming to ensure water resource safety and sustainability in response to climate change and human interference.

Emerging applications and ongoing challenges in WSN assisted IoT networks

In current scenario, Wireless Sensor Networks (WSN) based on Internet of Things (IoT) have seen a significant growth in research area. IoT is a network that enables sensors, and other events to interact with each other without human interference. In addition, wireless network is an essential part of the Internet of Things, generating an abundance of real-time applications. WS Ns can be recognized by their distributed sensor nodes and application fields including environmental monitoring, healthcare, agriculture, industrial automation, and smart cities. Therefore, an extensive usage of such networks is the outcome of their adaptability in variety of fields. The present study digs into the immense scope of WSN applications and the daunting challenges that come alongside. However, these applications face a slew of obstacles, including those related to energy, scalability, security, quality of service, network dependability, and data management. Further, the study focuses on gathering insights into the changing WSN technology environment by examining these applications and the issues they encounter, allowing to satisfy the demands of both existing and forthcoming deployments. As WSN technology is improving, the present study contributes by investigating prospective solutions, such as machine learning integration, energy harvesting, and standardization efforts.

Health assessment of the current situation of rivers and discussion on ecological restoration strategies: A case study of the Gansu province section of the Bailong river

The health of watershed ecology is mutually constrained by the sustainable development of the regional economy and society, and health assessment helps to identify the ecological status quo. As one of the “ethnic corridors”, the Bailong River has become a region of human solid interference. Therefore, this study constructed an indicator evaluation system to reflect rivers' natural attributes and functionality by analyzing candidate indicators' sensitivity, principal components, and independence. The weight of the indicators was determined using the Analytic Hierarchy Process combined with expert opinions. The ecological status of the Bailong River was evaluated through field monitoring data, and ecological restoration suggestions were proposed. The results show that Bailong River's comprehensive score is 58.27, indicating sub-health. The constructed evaluation system includes thirteen indicators, of which 53.84% are in an ideal state, 15.39% are healthy, 7.69% are sub-health, 15.39% are unhealthy, and 7.69% are pathological, with weak spatial heterogeneity. The evaluation results of the guidelines show that the state of “cleanliness” and “service” is ideal, “circulation” and “life” are healthy, and “physical structure” is worrying. Based on this, the ecological restoration and protection of the Bailong River in the future should focus on the following aspects: improving methods and theories, restoring the habitat environment of the river, ensuring water quantity, improving water quality, and conserving fish species. This study can provide necessary theoretical support for the ecological restoration of severely disturbed water-deficient rivers and lakes in northern China.

BUILDING A MOUSE BY EYEBALL MOVEMENT USING MACHINE LEARINING

An individual Human computer interference system is being introduced. In olden times, as an input device the mouse and keyboard were used by human computer interference system. Those people who are suffering from certain disease or illness cannot be able to operate computers. The idea of controlling the computers with the eyes will serve a great use for handicapped and disabled person. Also this type of control will eliminate the help required by other person to handle the computer. This measure will be the most useful for the person who is without hands through which they can operate with the help of their eye movements. The movement of the cursor is directly associated with the center of the pupil. Implementing a controlling system in it enables them to move without the help of another person is very helpful. First detect pupil center position of eye. Then the different variation on pupil position different command set for virtual keyboard. The signals pass the driver to interface with virtual keyboard itself. The driver will control both speed and direction to enable virtual keyboard to move forward, left, right and stop. Key Words: Human computer interference, Keyboard, Mouse, Eye, Cursor, pupil.

The overlooked margins: how cities impact diversity of plants and terrestrial invertebrates along urban streams

AbstractThe effect of different urbanization levels on terrestrial biodiversity associated with riparian zones remains poorly studied, despite the important ecosystem services it provides in cities. Studies focused on aquatic ecosystems in urban streams reported decreases in ecological quality and species richness, and lack of sensitive taxa. Thus, we investigated the impact of urbanization on riparian zone flora and terrestrial invertebrates using as case studies nine urban streams spanning an urbanization gradient in the Portuguese city of Coimbra. An unexpectedly high number of taxa were identified (163 plants, 80 terrestrial invertebrates), 80% new registrations for the region and one endemic-rare plant. Yet, diversity varied across streams, β-diversity reaching 39% Bray-Curtis similarity and averaging 25%, due to many underrepresented terrestrial taxa (N = 62 observed only in one site). Variation across the urbanization gradient reached 59% in plants and 82% in terrestrial invertebrates. High numbers of non-native taxa (13%; N = 32), mostly plants, suggested urbanization factors, e.g. human interference and discharge variation, may have favored invasions. Indeed, constructed natural spaces along streams, like parks, supported less biodiversity. Moreover, plant richness, especially of riparian plants, was correlated negatively with percentages of surrounding impervious areas and positively with water quality. This shows urbanization has negative effects on riparian vegetation, and water quality impacts both aquatic and terrestrial communities. Our results stress urban streams as novel ecosystems constituted by high numbers of non-native taxa, and the significance of riparian zones for biodiversity preservation. Less intense intervention on riparian vegetation is recommended to increase biodiversity.

Investigation and analysis of the community structure characteristics of zooplankton in Zibo

In order to understand the water ecological environment and the characteristics of plankton community structure in the Zibo area, a survey and study of plankton was conducted in the Zibo Basin in the autumn of 2022. By collecting zooplankton from nine major rivers within the country, a total of 122 species were collected from four phyla. Among them, rotifers are the most abundant, with a total of 77 species, accounting for 63%; Next are protozoa, with a total of 24 species, accounting for 20%; There are a total of 11 species of cladocerans, accounting for 9%; There are 10 species of copepods, accounting for 8%. The average density is 229.58/L, and the average biomass of zooplankton is 2.1997mg/L. The Shannon Wiener diversity index for zooplankton is 3.35. The survey results show that the water quality in the Zibo section is generally in a light pollution state. It is recommended to take further effective measures to strengthen the environmental governance around the watershed, reduce the discharge of domestic sewage, and reduce human activity interference.

UAS-based mapping of the July 3, 2019, ballistics density distribution on the W flank of Stromboli with uncertainty quantification

Stromboli is a volcanic island in a persistent state of activity, located off the northern coast of Sicily. The paroxysms have been the most powerful explosive phenomena at Stromboli in the last 150 years. These explosions can produce ballistic projectiles that can heavily affect trails and observation sites as well as hit inhabited areas at lower elevations, down to the coast. On July 3, 2019, a paroxysm significantly affected a great portion of the island with ballistic projectiles. In particular, many decimeter and meter-sized spatter bombs hit the W flank of Stromboli, and ignited multiple fires. In May 2022, we conducted an Unmanned Aerial System photogrammetric campaign over a sector of the W flank of Stromboli that was heavily affected by the paroxysm. The largest clasts were still preserved after 3 years, not disturbed by significant mass wasting phenomena or human interference, and they were not yet hidden by the post-fire regrowth of the brush vegetation. In this study, we constrained the main sources of uncertainty affecting the bombs distribution on the ground, by characterizing a percentage of uncertain clasts, testing various density estimators, and by modeling an areal buffer around the mapped clasts. We produced a 0.18 km2 wide 1.6-cm resolution orthomosaic, a 10-cm resolution Digital Surface Model, and 2813 outlines of the mapped ballistics. Spatial distribution of the ground cover and associated uncertainty were analyzed as a function of the distance and of the angular direction from the source.

Spatial distribution characteristics of soil heavy metals in Sabao Chaqu watershed of Tuotuo river, Qinghai-Tibet Plateau based on geographic detector

The Qinghai-Tibet Plateau belongs to the area of extremely fragile environment and sensitive to human activities. In recent years, more and more human interference has been detected in this area. In this study, 128 surface soil samples were collected from the Sabao Chaqu watershed of the Tuotuo river at the source of the Yangtze River on the Qinghai-Tibet Plateau. The soil pollution status and spatial distribution characteristics of Cd, Hg, As, Cu, Pb, Cr, Zn and Ni were evaluated by soil accumulation index, enrichment factor, pollution index and geographical detector. The results showed that the average contents of As, Cd, Pb and Zn in the study area were 1.2–3.64 times higher than soil background values of Tibet, while the contents of Hg, Cr, Cu and Ni were lower than the background values, while the average content of As was higher than the soil pollution risk screening value (GB15618-2018), and the pollution index showed that As was in a low pollution state, while the other 7 heavy metals were in a safe state. There were significant differences in the spatial distribution of 8 heavy metals and there was a significant correlation with soil properties and distance factors. Factor detection showed that natural factors had the strongest explanatory power to the contents of As, Cd, Cr, Cu and Ni, distance from the lake and soil Sc content had the strongest explanatory power to Hg content, and anthropogenic factors had the strongest explanatory power to Pb content. Interaction detection revealed that the q values of the strongest interaction explanatory power for As, Cd, Cr, Cu, Hg, Ni, Pb, and Zn were 2.81, 4.30, 1.26, 2.47, 2.33, 1.59, 6.37, and 5.08 times higher than their strongest factor detection explanatory power, respectively. The interaction between anthropogenic factors and other factors has an important influence on the spatial differentiation of heavy metals in the study area. Risk detection showed that the average contents of As, Cd, Cr, Cu, Hg, Ni, Pb and Zn were the highest in the subregions of MgO, TS, Sc, X6, X13, MgO, TN and X4, respectively. Comprehensive study shows that the spatial differentiation of As, Cd, Cr, Cu, Ni and Zn is mainly affected by natural factors, but there are also some anthropogenic factors, the spatial differentiation of Hg is affected by both natural factors and atmospheric deposition, and the spatial distribution characteristics of Pb are mainly affected by anthropogenic factors.

River ecosystem health assessment in the Qinghai-Tibet Plateau: A novel hybrid method based on artificial intelligence and multi-source data fusion

River ecosystem health assessment (REHA), an effective approach for identifying river ecosystem health, is crucial for achieving sustainable river management and ensuring water security. However, existing REHA methods still fail to consider the cumulated influences of uncertain inputs, stochastic environment and limited rationality of decision makers on REHA. Additionally, current REHA studies have mainly concentrated on plain areas, while the Qinghai-Tibet Plateau (QTP) remains largely unknown. Developing REHA techniques for plateau rivers is an urgent matter, due to the heightened fragility and complexity of river ecosystems in the QTP. To accurately assess river ecosystem health in the QTP, this study proposed Pythagorean fuzzy cloud (PFC) via coupling the Pythagorean fuzzy sets and cloud model. A novel PFC-TODIM model was developed by extending TODIM (the acronym in Portuguese for interactive and multicriteria decision making) to the Pythagorean fuzzy environment. The hybrid decision making framework was then created to handle REHA with uncertain inputs and stochastic environment, and the Senge Tsangpo River (STR) served as a case study in the QTP. We developed the indicator system based on multi-source data fusion, and employed Bayesian model averaging (BMA) method to reveal the potential risks and driving factors of river ecosystem health. Results showed that the developed models considered the limited rationality of decision makers, effectively handled REHA with uncertainties, and avoided overestimating river health levels due to ignoring the randomness and fuzziness of REHA. In STR, health statuses exhibited marked spatial differences. Sampling sites of 9.091%, 77.273 % and 13.636 % were excellent, healthy and subhealthy, respectively. Our findings highlight that dams, urban development, fish release, and grazing have adverse impacts on STR health, and effective protection measures are required to minimize human interferences on ecologically fragile areas. These findings can improve our understanding of the human disturbances and natural factors that interfere with river ecosystem health in the QTP.

Genomic Diversity and Demographic History of the Endangered Sichuan hill-partridge (Arborophila rufipectus).

Species conservation can be improved by knowledge of genetic diversity and demographic history. The Sichuan hill-partridge (Arborophila rufipectus, SP) is an endangered species endemic to the mountains in southwestern China. However, little is known about this species' genomic variation and demographic history. Here, we present a comprehensive whole-genome analysis of six SP individuals from the Laojunshan National Nature Reserve in Sichuan Province, China. We observe a relatively high genetic diversity and low level of recent inbreeding in the studied SP individuals. This suggests that the current population carries genetic variability that may benefit the long-term survival of this species, and that the present population may be larger than currently recognized. Analyses of demographic history showed that fluctuations in the effective population size of SP are inconsistent with changes of the historical climate. Strikingly, evidence from demographic modeling suggests SPs population decreased dramatically 15,100 years ago after the Last Glacial Maximum (LGM), possibly due to refugial isolation and later human interference. These results provide the first detailed and comprehensive genomic insights into genetic diversity, genomic inbreeding levels, and demographic history of the Sichuan hill-partridge, which are crucial for the conservation and management of this endangered species.

A convolutional neural network based on an evolutionary algorithm and its application

PM2.5 concentration predictions can provide air pollution control, management, and early warning. However, the PM2.5 data with high-dimensionality, complexity, and dynamics pose a great challenge to achieve optimal prediction results. Convolutional Neural Networks (CNNs) has unique advantages in processing complex data and contributes to state-of-the-art performances. However, designing the architecture and selecting the learning rate for CNN are time-consuming and requires prior knowledge. Evolutionary algorithms, with the advantages of global convergence, ergodicity, robustness and adaptability, are the most commonly used methods to design the optimal framework for CNNs. Therefore, to improve the predictive performance of CNNs, this paper proposes an improved CNN method (EBRO-ICNN) which employs the enhanced battle royale optimization (EBRO) algorithm and proportional-derivative (PD) control. Firstly, the EBRO algorithm with multistrategy collaborative optimization is introduced, and validated by CEC2017 benchmark functions, which demonstrates EBRO strong global exploration capability, fast convergence speed, and low time complexity. Next, PD control is applied to adjust the learning rate of CNN (ICNN) dynamically, which improves the efficiency and stability of the network training process. At the same time, the ICNN model is optimized using the EBRO algorithm, which can reduce the human interference, generate the optimal framework automatically and enhance the prediction accuracy effectively. Finally, a private air quality dataset and two public datasets are utilized to evaluate the performance of the EBRO-ICNN model, considering 3 error evaluation metrics and 7 prediction comparison models. The experimental results demonstrate that the EBRO-ICNN model exhibits great accuracy and stability.