I. Introduction

With global climate change and increasing human activities, biodiversity is facing unprecedented threats. As an emerging means of observation characterized by high precision, high efficiency and high automation, NEOP provides new opportunities for biodiversity research.

The SCI-in-One launched by Beijing StarView Technology Co., Ltd. is a comprehensive near-Earth observation platform, which has the function of integrating observation of vegetation, meteorology, soil, hydrology and other factors, and realizes the mode of multiple applications of one set of system.

Through the research and development of StarView's scientific researchers, the function of One Pole Pass is now expanded to the observation of biodiversity through the integration of multiple technologies. By integrating environmental factors, multi-spectral and acoustic technologies, it has constructed an integrated biodiversity observation system for near-Earth observation platforms, which can already realize comprehensive, systematic and continuous observation of biological species such as plants, mammals and birds, and provide a scientific basis for biodiversity protection.

The expansion of the function of the product is not only to promote the in-depth development of biodiversity research, but also to provide a scientific basis for biodiversity conservation. Through the construction of an integrated biodiversity observation system based on the near-Earth observation platform, we can gain a more in-depth understanding of the distribution, quantity, dynamic changes and other information of biodiversity, which will provide strong support for the formulation of more scientific and effective conservation measures. At the same time, this study can also provide useful reference for research in other fields and promote the continuous innovation and development of scientific research.

II. Objects of observation

2.1 Plants

In plant observation, the application of intelligent multispectral cameras provides us with a new perspective. Such cameras are able to capture the reflective properties of plants in different spectral bands, thus revealing the physiological states and biochemical processes within the plant. For example, by monitoring the red-edge characteristics of plant leaves, we can assess the photosynthetic efficiency and water status of plants, which is important for understanding the response mechanism of plants to environmental changes.

In the process of plant observation, it is also necessary to pay attention to the interaction between plants and the environment. For example, the response mechanism of plants to environmental factors such as soil nutrients, water and light, and the multi-parameter observation function of the scientific research one-shot pass provides the possibility of realizing the integrated observation of the above parameters. Based on these observations, we can better understand the role and position of plants in the ecosystem and provide scientific basis for biodiversity conservation and ecological restoration.

2.2 Mammals

In the observational study of mammals, a variety of technical methods have been used to comprehensively reveal their ecological habits and behavioral characteristics. First, intelligent infrared cameras have been used to monitor the activity patterns of mammals over time; by analyzing the captured images, significant differences can be found in the activity time, range of activity and food selection of different species. For example, certain nocturnal mammals are most active at dusk and dawn, while daytime-active mammals are more secretive during these hours. It was also found that the food choices of different species were also influenced by environmental factors, such as seasonal variations and the abundance and distribution of food resources.

In terms of mammal sound recording, the system also adopts voiceprint technology. By collecting and analyzing the calls of different species, we have established a rich voiceprint library and realized the automatic recognition of animal sounds. The application of this technology not only improves the efficiency of data processing, but also provides us with more information about animal behavior. For example, by analyzing the frequency, duration and change patterns of the calls, we can infer the sex, age and emotional state of the animals.

In order to gain a deeper understanding of the ecological adaptations of mammals, the system combines big data and machine learning techniques to analyze the observation data in depth. Through the establishment of species identification models, quantitative statistical models and behavioral analysis models, the laws of mammal population dynamics, habitat selection and interspecific relationships can be revealed. The establishment of these models not only improves our understanding of mammal ecology, but also provides a scientific basis for ecological protection and management. 

2.3 Birds

Soundprint technology is able to realize automatic identification and behavioral analysis of bird species by recording and analyzing the sound characteristics of birds. Through the soundprint technology, we have established a bird soundprint library to digitally store and compare the sound characteristics of different bird species, thus realizing the rapid identification of bird species. In addition, soundprint technology can be used to analyze the behavioral patterns of birds, such as the frequency, duration, and rhythm of chirping sounds, thus revealing the habits and migration patterns of birds.

In addition to acoustic technology, the NEOP can also combine other observation means, such as infrared cameras for multi-scale observation of birds. These observation tools can provide rich data support and help us understand the living habits and migration patterns of birds more comprehensively. For example, through infrared cameras, we can record the activities of birds at night, thus revealing the adaptation mechanism of birds in the dark environment.

In bird research, we can also draw on the theories and methods of ecology, behavior and other multidisciplinary disciplines to conduct an in-depth analysis of the interaction between birds and the environment. For example, we can apply ecological theories to analyze the distribution and quantitative changes of birds in different ecological environments, so as to reveal the effects of ecological environments on bird populations. Meanwhile, we can also apply behavioral theories to study the behavioral patterns and adaptive mechanisms of birds, so as to provide scientific basis for bird conservation and ecological restoration.

III. Instrument configuration

3.1 Intelligent Multispectral Camera

The system is equipped with a multispectral climate camera with a pan-tilt zoom, a device that can capture the subtle changes of plants in different spectral bands, thus revealing the growth status, health of plants and their interaction with the environment. Intelligent multispectral cameras are also characterized by high temporal and spatial resolution, enabling continuous monitoring of dynamic changes in plant communities. Through the time-series image data, plant climate models can be constructed to analyze the growth cycle and rhythmic changes of plants, and then predict their response to climate change.

3.2 Sound Recorder

The principle of voiceprint technology, as a biometric identification technology, is based on the uniqueness and stability of voice signals. Every human voice, and even every animal voice, contains unique features such as frequency, pitch and rhythm, which form the basis of voiceprint. In animal sound recognition, soundprint technology is able to achieve accurate recognition of animal species by capturing and analyzing these sound features.

Voiceprint technology can also be used for animal behavior analysis. By analyzing the changes in animals' voices, we can understand their behavioral patterns, emotional states, and changes in their ecological environment. For example, when animals are in a state of nervousness or fear, their voices may change, and such changes can be detected and analyzed by acoustic pattern technology. This is of great significance for studying the ecological habits of animals and protecting their ecological environment.

IV. Software platform

4.1 Plant species identification module

In the Integrated Biodiversity Observing System, the plant species recognition module based on big data machine learning plays a crucial role. The module utilizes massive image data to realize automatic identification and classification of plant species through deep learning algorithms. Compared with traditional plant classification methods, the plant species recognition module based on big data machine learning has higher accuracy and efficiency.

The module first acquires high-definition images of plants through an intelligent multispectral camera, and then utilizes image preprocessing techniques to perform operations such as denoising and enhancement to improve the image quality. Then, the pre-processed images are subjected to feature extraction and classification by means of a trained deep learning model. Finally, based on the classification results, corresponding species identification reports are generated to provide strong data support for ecologists and botanists.

4.2 Automatic fruit counting module

Based on advanced image processing and machine learning algorithms, it is able to realize fast and accurate counting of the number of plant fruits. By introducing a deep learning model, the module is able to recognize and distinguish different kinds of fruits, and then realize automatic counting. In practical applications, the module has achieved remarkable results.

By accurately counting the number of plant fruits, researchers can gain an in-depth understanding of plant reproductive strategies and population dynamics, providing a scientific basis for ecological protection and restoration. At the same time, the module can also provide decision support for agricultural production and forestry management and promote sustainable agricultural development.

4.3 Greenness index time series analysis module

The time-series analysis module of the plant greenness index is an important part of the NEOMAP Integrated Biodiversity Observing System (IBS). By analyzing the time series of plant greenness index, this module can reveal the growth status of plants, ecological changes and ecosystem health. In the process of data analysis, we adopted the time series analysis method, and carried out smoothness test, seasonality analysis and other operations on the time series of plant greenness index. Through these analyses, we were not only able to understand the trend of the plant greenness index, but also able to predict the growth condition of plants in the future period. In addition, we also used regression analysis and other methods to further explore the relationship between plant greenness index and environmental factors, which provides a scientific basis for the management and protection of ecosystems.

4.4 Module for Automatic Recording and Recognition of Flowering Periods

Flowering period, as a key stage in the plant life cycle, not only affects plant reproduction and survival, but is also an important indicator for assessing the health of ecosystems. Therefore, accurate and efficient recording and identification of plant flowering periods is important for understanding plant ecology and conserving biodiversity.

The development of the automatic plant flowering period recording and recognition module benefits from the rapid development of big data and machine learning technology. By constructing a large-scale plant image database and combining it with deep learning algorithms, we can train efficient plant flowering stage recognition models. These models can automatically analyze plant images and accurately identify the flowering stage of plants, thus greatly improving the observation efficiency and accuracy.

In addition, the application of the module for automatic recording and identification of flowering period is not only limited to field observation. In horticulture, agriculture and other fields, the module also has a wide range of application prospects. For example, in horticulture design, we can use the module to predict the flowering time of flowers, so as to rationally arrange the planting and display of flowers. In agricultural production, the module can help farmers keep abreast of the growth of crops, providing a decision-making basis for precision agricultural management.

4.5 Automatic animal species identification module based on voiceprint technology

Using advanced acoustic pattern recognition technology, efficient and accurate identification of animal species is achieved by capturing and analyzing the unique characteristics of animal sounds. Compared with traditional methods of animal species identification, voiceprint recognition not only improves the accuracy of identification, but also greatly enhances the efficiency and scope of observation.

The automatic animal species recognition module based on voiceprint technology has established a huge voiceprint database by collecting and analyzing a large amount of animal voice data. This database contains the sound characteristics of various animals, which provides a solid foundation for subsequent voiceprint recognition. At the same time, the module also adopts advanced machine learning algorithms to improve the accuracy and stability of recognition through continuous learning and optimization of the voiceprint data.

4.6 Animal Rhythm Analysis Module Based on Voiceprint Technology

In terms of the rhythm analysis of animal species, the module based on acoustic pattern technology can realize the continuous recording of animal sound signals for a long period of time, and then analyze the activity rhythms, breeding behaviors and other important ecological information of animals. For example, through the recording and analysis of bird calls, it can reveal the changes in the frequency and intensity of birds' calls in different seasons and time periods, thus inferring their ecological behaviors such as breeding and migration.

The animal species rhythm analysis module based on voiceprint technology is also highly automated and intelligent. Through the training and optimization of machine learning algorithms, the module is able to realize automatic classification and recognition of animal sound signals, greatly improving the accuracy and efficiency of observation. At the same time, the module is also able to combine other observation data, such as environmental temperature and humidity, to comprehensively analyze the relationship between the rhythmic changes of animal species and environmental factors, providing more in-depth and comprehensive data support for ecological research.

V. Conclusion

We will continue to deepen the application of near-Earth observation platforms in the Integrated Biodiversity Observing System. On the one hand, we will work to enhance the intelligence and automation of observation equipment, such as optimizing the performance of intelligent multispectral cameras and acoustic technology, in order to improve the accuracy and efficiency of observation data. On the other hand, we will explore more kinds of biological observation objects, such as insects and reptiles, to gain a more comprehensive understanding of the distribution and changes of biodiversity. In addition, we will also focus on the interactions and impacts among ecosystems to reveal the deep-rooted causes of biodiversity changes.

Looking to the future, we expect to build a more comprehensive integrated biodiversity observation system through continuous research and innovation. This system will be able to monitor and early-warning changes in biodiversity in real time and provide scientific basis for ecological protection and environmental governance. At the same time, we also look forward to promoting the in-depth development of biodiversity research through interdisciplinary cooperation and exchanges, and contributing to the sustainable development of mankind.

  E.O. Wilson, a famous ecologist, said that biodiversity is the cornerstone of life on earth and the foundation of our survival and development. We will continue to devote ourselves to the research of the Integrated Biodiversity Observing System (IBOS), responding to the national sustainable development concept of "green mountains are golden mountains", and contributing to the protection of the Earth's biodiversity and the promotion of the construction of an ecological civilization.