A Computer Vision System for the Automatic Classification of Five Varieties of Tree Leaf Images

A computer vision system for automatic recognition and classification of five varieties of plant leaves under controlled laboratory imaging conditions, comprising: 1–Cydonia oblonga (quince), 2–Eucalyptus camaldulensis dehn (river red gum), 3–Malus pumila (apple), 4–Pistacia atlantica (mt. Atlas mastic tree) and 5–Prunus armeniaca (apricot), is proposed. 516 tree leaves images were taken and 285 features computed from each object including shape features, color features, texture features based on the gray level co-occurrence matrix, texture descriptors based on histogram and moment invariants. Seven discriminant features were selected and input for classification purposes using three classifiers: hybrid artificial neural network–ant bee colony (ANN–ABC), hybrid artificial neural network–biogeography based optimization (ANN–BBO) and Fisher linear discriminant analysis (LDA). Mean correct classification rates (CCR), resulted in 94.04%, 89.23%, and 93.99%, for hybrid ANN–ABC; hybrid ANN–BBO; and LDA classifiers, respectively. Best classifier mean area under curve (AUC), mean sensitivity, and mean specificity, were computed for the five tree varieties under study, resulting in: 1–Cydonia oblonga (quince) 0.991 (ANN–ABC), 95.89% (ANN–ABC), 95.91% (ANN–ABC); 2–Eucalyptus camaldulensis dehn (river red gum) 1.00 (LDA), 100% (LDA), 100% (LDA); 3–Malus pumila (apple) 0.996 (LDA), 96.63% (LDA), 94.99% (LDA); 4–Pistacia atlantica (mt. Atlas mastic tree) 0.979 (LDA), 91.71% (LDA), 82.57% (LDA); and 5–Prunus armeniaca (apricot) 0.994 (LDA), 88.67% (LDA), 94.65% (LDA), respectively.

Chemical Composition and Acaricidal Activity of the Essential Oils of Some Plant Species of Lamiaceae and Myrtaceae against the Vector of Tropical Bovine Theileriosis: Hyalomma scupense (syn. Hyalomma detritum)

The present study aimed to investigate the acaricidal properties of six essential oils. They were extracted from some plant species (Lamiaceae and Myrtaceae) using the technique of hydrodistillation with the Clevenger apparatus. The chemical compositions of the essential oils under study were determined by gas chromatography–mass spectrometer (GC-MS). An Adult Immersion Test (AIT) and a Larval Immersion Test (LIT) were used to evaluate the acaricidal activity of these essential oils against the adults and larvae of Hyalomma scupense. GC-MS analysis showed the major constituents of each essential oil: 25.49% of α-thujone (lavender); 46.82% of carvacrol (oregano); 78.78% of carvacrol (thyme); 40.27% of 1,8-cineole (blue gum); 17.45% of p-cymene (river red gum); and 26.96% of 1,8-cineole (rosemary). The biotests on the essential oils revealed that they inhibit the reproduction of H. scupense engorged females at a rate of 100 % with doses of 0.781 μl/ml of rosemary, 1.562 μl/ml of thyme, 3.125 μl/ml of lavender and oregano, and 6.250 μl/ml of blue gum and river red gum. After a treatment that lasted for 24 hours, essential oils showed a larvicidal activity with respective values of lethal concentrations (LC): LC50, LC90, and LC95 (0.058, 0.358, and 0.600 μl/ml for thyme; 0.108, 0.495, and 0.761 μl/ml for rosemary; 0.131, 0.982, and 1.740 μl/ml for oregano; 0.155, 2.387, and 5.183 μl/ml for blue gum; 0.207, 1.653, and 2.978 μl/ml for river red gum; and 0.253, 2.212, and 4.092 μl/ml for lavender). This is the first report on the acaricidal activity of these essential oils against H. scupense. The results obtained showed that the essential oils with chemotype carvacrol, 1,8-cineole, α-thujone, and p-cymene are highly acaricidal, and they can be used for ticks control. However, further studies on their toxicity in nontarget organisms are required.

Phosphorus speciation and dynamics in river sediments, floodplain soils and leaf litter from the Lower Murray River region

Phosphorus (P) availability, which depends on both P concentration and speciation, often controls primary productivity and algal-bloom formation in river systems. The river P pool is also connected to P pools of adjacent sediments, soils and vegetation. Thus, informed management of P in floodplain–river systems requires detailed understanding of P concentration and speciation in all of these interconnected components. We studied P speciation in river sediments and water, floodplain soils and river red gum (Eucalyptus camaldulensis) leaf litter from the Lower Murray region using conventional spectroscopic measurements, solution 31P nuclear magnetic resonance (31P NMR) spectroscopy, and leaching experiments to simulate floodplain re-wetting of leaf litter. Almost all (>85%) of the P in river sediments was in the orthophosphate form, whereas floodplain soils had higher proportions of organic P (PO) species. Both fresh and senescent river red gum leaf litter also had a much higher concentration of PO, primarily in the form of phytate. On submersion, there was a rapid (0–96h) loss of dissolved P from senescent leaves; release of dissolved organic carbon showed similar kinetics. Loss of P from the leaves included both organic and inorganic forms. The results have important implications for aquatic primary productivity and environmental management strategies.

Experimental evaluation of the initial effects of large-scale thinning on structure and biodiversity of river red gum (Eucalyptus camaldulensis) forests

Context Multi-use management of global forests has seen even-aged, high-stem density regrowth represent >50% of the world’s forest cover. Large areas of river red gum (Eucalyptus camaldulensis) forests have declined in ecological condition. Thinning has been promoted as a tool to reduce competition in dense, young stands of E. camaldulensis regrowth, yet responses of forest structure and fauna to large-scale thinning are largely unclear. Aims To establish a before-after-control-impact experiment to assess responses of forest structure and fauna to large-scale (compartment-level; ~440 ha) silvicultural thinning. Methods We measured immediate (<2 yrs) responses of forest structural components (living, dead and hollow-bearing stem densities, coarse woody debris (CWD) density and volumes and ground cover) and components of biodiversity (bats, birds, volant insects and non-volant mammals) before and after thinning within five control and three impact compartments. Key results Thinning reduced stem density by approximately two-thirds and was associated with a substantial increase in activity and richness of bats and a change in bat species composition. There was no change in richness for birds and non-volant mammals, nor insect biomass in relation to thinning. However, thinning affected composition of non-volant mammals, with the common brushtail possum (Trichosurus vulpecula) and red fox (Vulpes vulpes) less active post-thinning at impact plots relative to control plots. Thinning reduced the density of dead stems, though these were predominantly small (~13 cm diameter at breast height over bark, or dbhob) and mostly lacked hollows. Hollow-bearing tree density was not affected by thinning. Although thinning increased CWD densities, volume of CWD did not change, indicating that thinning contributed small-sized CWD. Thinning did not affect densities of hollow- and fissure-bearing CWD or ground cover, which was low (<7%) in control and impact plots. There were no other negative effects on biodiversity detected. Conclusions Short-term fauna responses to thinning were generally neutral or positive. Implications Ongoing monitoring is required to detect long-term changes that may result from colonisation or altered breeding success after thinning. We recommend that some unthinned stands should be retained throughout the landscape to provide a mosaic forest structure suitable for a diverse fauna.