Plant Species Identification based on Plant Leaf Using Computer Vision and Machine Learning Techniques

Abstract

Plants are very crucial for life on Earth. There is a wide variety of plant species available, and the number is increasing every year. Species knowledge is a necessity of various groups of society like foresters, farmers, environmentalists, educators for different work areas. This makes species identification an interdisciplinary interest. This, however, requires expert knowledge and becomes a tedious and challenging task for the non-experts who have very little or no knowledge of the typical botanical terms. However, the advancements in the fields of machine learning and computer vision can help make this task comparatively easier. There is still not a system so developed that can identify all the plant species, but some efforts have been made. In this study, we also have made such an attempt. Plant identification usually involves four steps, i.e. image acquisition, pre-processing, feature extraction, and classification. In this study, images from Swedish leaf dataset have been used, which contains 1,125 images of 15 different species. This is followed by pre-processing using Gaussian filtering mechanism and then texture and color features have been extracted. Finally, classification has been done using Multiclass-support vector machine, which achieved accuracy of nearly 93.26%, which we aim to enhance further.

Genetic, morphological, and niche variation in the widely hybridizingRhus integrifolia‐Rhus ovataspecies complex

“…For example, using the application iNaturalist (https://www.inaturalist.org/), there are 170 records of R. integrifolia × ovata (likely to be an underestimate of actual abundance), 5715 of R. integrifolia and 3591 of R. ovata (last accessed April 14, 2020). Although these types of observations have obvious biases (e.g., clustered near population centers or public lands; for example Dickinson et al, 2012), if properly verified by an expert or via machine learning (e.g., Kaur & Tulsi, 2019; Priya, Balasaravanan, & Thanamani, 2012; Wilf et al, 2016) they may ultimately improve distribution models for hybridizing species, including R. integrifolia and R. ovata .…”

Reconhecimento de Plantas Medicinais Usando Características de Cor, Textura e Forma

“…Em [Chaki et al 2019], características de cor e textura das bordas de folhas são combinadas com o objetivo de fazer a identificação de imagens de folhas fragmentadas. Em [Kaur and Tulsi 2019], a forma e cor das folhas também são utilizadas. Em [Yigit et al 2019], em adiçãoà forma eà cor, a dimensão e a textura também são extraídas, além de padrões obtidos através das coordenadas e do centro das folhas.…”

“…Dentre os principais algoritmos e abordagens podemos citar Máquina de Vetores de Suporte [Kaur andTulsi 2019, Yigit et al 2019], Redes Neurais Artificiais [Sun et al 2017, Chaki et al 2019, Kaya et al 2019, k-Vizinhos Mais Próximos [Pearline et al 2019, Yigit et al 2019] e o Classificador de Floresta Aleatória [Pearline et al 2019, Yigit et al 2019. Como o foco do presente trabalhoé na utilização de características das cores, da textura e da forma das folhas, extraem-se primeiro as característica das cores dos segmentos de imagens (que estão no padrão RGB), considerando-se um canal do RGB por vez.…”

Short Communication: Digital identification approach to characterize Hevea brasiliensis leaves

“…In manual identification, special plant characteristics are determined as those that help in identifying the intended plant species. Conventional plant species identification methods (Kaur and Kaur 2019) seem impractical to ordinary people and are a challenge for professional taxonomy as well. Manual identification is often time-consuming and inefficient.…”

“…tree species need to be set up to preserve the type of clones and as a source of elders for breeding programs. Introduction of plants through leaf shapes using image processing as initial PRATOMO et al -Digital identification of Hevea brasiliensis 1007 processing inputted leaf images (Kaur and Kaur 2019) converted into binary images will then be further processed to find features (Wirdiani and Sudana 2016). Many methods can be used to calculate the extraction of features as well as the introduction of plants by identifying the shape (Agus et al 2015, Pallavi andDevi 2014), segment (Lee and Hong 2013), color (Trishen et al 2015), and leaf texture (Bhardwaj et al 2013;Kadir et al 2012).…”