An Analysis of the Accuracy of Photo-Based Plant Identification Applications on Fifty-Five Tree Species

Background: With the creation of photo-based plant identification applications (apps), the ability to attain basic identifications of plants in the field is seemingly available to anyone who has access to a smartphone. The use of such apps as an educational tool for students and as a major identification resource for some community science projects calls into question the accuracy of the identifications they provide. We created a study based on the context of local tree species in order to offer an informed response to students asking for guidance when choosing a tool for their support in classes. Methods: This study tested 6 mobile plant identification apps on a set of 440 photographs representing the leaves and bark of 55 tree species common to the state of New Jersey (USA). Results: Of the 6 apps tested, PictureThis was the most accurate, followed by iNaturalist, with PlantSnap failing to offer consistently accurate identifications. Overall, these apps are much more accurate in identifying leaf photos as compared to bark photos, and while these apps offer accurate identifications to the genus-level, there seems to be little accuracy in successfully identifying photos to the species-level. Conclusions: While these apps cannot replace traditional field identification, they can be used with high confidence as a tool to assist inexperienced or unsure arborists, foresters, or ecologists by helping to refine the pool of possible species for further identification.

Medicinal Plant Identification Using Machine Learning Techniques

In this world there are thousands of plant species available, and plants have medicinal values. Medicinal plants play a very active role in healthcare traditions. Ayurveda is one of the oldest systems of medicinal science that is used even today. So proper identification of the medicinal plants has major benefits for not only manufacturing medicines but also for forest department peoples, life scientists, physicians, medication laboratories, government, and the public. The manual method is good for identifying plants easily, but is usually done by the skilled practitioners who have achieved expertise in this field. However, it is time consuming. There may be chances to misidentification, which leads to certain side effects and may lead to serious problems. This chapter focuses on creation of image dataset by using a mobile-based tool for image acquisition, which helps to capture the structured images, and reduces the effort of data cleaning. This chapter also suggests that by ANN, CNN, or PNN classifier, the classification can be done accurately.

Biological Trace Information Extracted from Bioaerosols Using NGS Analysis

Bioaerosols are atmospheric particles with a biological trace, such as viruses, bacteria, fungi, and plant material such as pollen and plant debris. In this study, we analyzed the biological information in bioaerosols using next generation sequencing of the trace DNA. The samples were collected using an Andersen air sampler and separated into two groups according to particulate matter (PM) size: small (PM2.5) and large (PM10). Amplification and sequencing of the bacterial 16S rDNA gene, prokaryotic internal transcribed spacer 1 (ITS1) region and DNA sequence of a plant chloroplast gene (rbcL) were carried out using several sets of specific primers targeting animal and plant sequences. Lots of bacterial information was detected from the bioaerosols. The most abundant bacteria in several samples were of the Actinobacteria (class), Alphaproteobacteria, Bacilli, and Clostridia. For the animal detection using internal transcribed spacer 1, only uncultured fungi were detected in more than half of the hits, with a high number of Cladosporium sp. in the samples. For the plant identification, the ITS1 information only matched fungal species. However, targeting of the rbcL region revealed diverse plant information, such as Medicago papillosa. In conclusion, traces of bacteria, fungi, and plants could be detected in the bioaerosols, but not of animals using our primers.

Plant leaf identification system using convolutional neural network

This paper proposes a leaf identification system using convolutional neural network (CNN). This proposed system can identify five types of local Malaysia leaf which were acacia, papaya, cherry, mango and rambutan. By using CNN from deep learning, the network is trained from the database that acquired from leaf images captured by mobile phone for image classification. ResNet-50 was the architecture has been used for neural networks image classification and training the network for leaf identification. The recognition of photographs leaves requested several numbers of steps, starting with image pre-processing, feature extraction, plant identification, matching and testing, and finally extracting the results achieved in MATLAB. Testing sets of the system consists of 3 types of images which were white background, and noise added and random background images. Finally, interfaces for the leaf identification system have developed as the end software product using MATLAB app designer. As a result, the accuracy achieved for each training sets on five leaf classes are recorded above 98%, thus recognition process was successfully implemented.

Ayurvedic Plant Identification using Image Processing and Artificial Intelligence

Wireless networks provide small sensing, machine and wireless networking nodes. Different designs and implementation techniques were built based on the device requirements for wireless network sensors (WSN). Sensor networks are used in various applications, such as environmental monitoring, home automation, military applications, etc. In this study introduce an architectural survey and deployment of nodes in the Wi-Fi Sensor network in this article. The environmental features that can be added to the sensor networks are given. The program relies on the node installed in the WSN and is deterministic or random. But the biggest issue in both cases is the coverage of the region involved. Researcher also describe WSN routing protocols. In this paper, a new technique to deployment problem is proposed based on the artificial bee colony (ABC) algorithm which is enhanced for the deployment of sensor networks to gain better performance by trying to increase the coverage area of the network and energy consumption. The good performance of the proposed EABC algorithm shows that it can be utilized in the deployment of WSN.

The Morphological and Anatomical Studies of The Aerial Parts of Abroma augusta L. from Semarang

Abroma augusta L. known as Devil’s cotton belongs to Malvaceae. The exploratory study aimed to study the morphological and anatomical characteristics of the aerial parts of A. augusta L. from Semarang. The transverse section of the aerial parts was made by a simple method (fresh preparation) and then observed under a binocular microscope with an optilab. All characteristics were observed and then compared with the references. The collected data were analyzed descriptively and quantitatively. In summary, the results showed that A. augusta L. was an evergreen shrub (small tree) with orthotropic and plagiotropic branches and polymorphous leaves. The inflorenscence was found in the terminal and axillar plagiotropic branching with bisex, actinomorphic, and pentamerous flowers. The fruit was unique (obconical capsule with a rounded base and truncate-tip with 5 angled wings) including cotton fibers and numerous black seeds. The petiole was composed of epidermis, collenchyma, cortical parenchyma, sclerenchyma, vascular bundle, mucilaginous ducts, and pith. The dorsiventral leaf was composed of upper and lower epidermis, palisade, and spongy parenchyma. The stomata type was ranunculaceous (anomocytic) while the guard cell was kidney-shaped. The stomata density on the abaxial leaf was higher than that of the adaxial leaf. The stellate and unicellular non-glandular trichomes, and capitate glandular trichomes were found abundantly on the petiole and leaf blade. These morphological and anatomical studies are important to support the identification as a part of the conservation effort of the plant. Further studies are recommended to investigate the root morphology and anatomy and also biochemical characteristics of each part of the plant in order to obtain complete plant identification.

Variation of kelulut (Heterotrigona itama) habitat landscapes in South Kalimantan

Honey production depends on the availability of the landscape as a habitat for producing bee’s food sources. The purpose of this study was to determine different landscapes as a habitat for kelulut (Heterotrigona itama) bees in producing honey from 5 different stingless bee locations. The research was conducted in three districts: Hulu Sungai Tengah, Hulu Sungai Selatan and Tapin District, South Kalimantan Province. This research was conducted to record the types of vegetation in each landscape, which can be divided into three categories; 1 location was a combination type of forest and garden (type 1), 2 locations was a combination type of settlement, shrub, and paddy fields (type 2), and 1 location was a combination type of settlement, plantation, and shrub (type 3). Each meliponiculture also recorded the honey production every month. The results showed that the farmers’ number of beehives was between 96 and 252 hives/farmer. The average production in the rainy season is 0.17 L hive-1year−1, and the dry season is 0.24 L hive−1year−1. Honey production per year for each location was as follows: location type 1 produces 1.59 L hive−1, location type 2 produces 1.85 L hive−1, and location type 3 produces 2.41 L hive−1. Plant identification results at each type of location showed that the number of species found at vegetation cover type 1, 2, and 3 was 116, 128, and 107 species, respectively. At the farms with vegetation cover types 2 and 3, many different flowering shrubs provide year-round forage for the stingless bee.