Leaf recognition using BP-RBF hybrid neural network

Plant recognition has great potential in forestry research and management. A new method combined back propagation neural network and radial basis function neural network to identify tree species using a few features and samples. The process was carried out in three steps: image pretreatment, feature extraction, and leaf recognition. In the image pretreatment processing, an image segmentation method based on hue, saturation and value color space and connected component labeling was presented, which can obtain the complete leaf image without veins and background. The BP-RBF hybrid neural network was used to test the influence of shape and texture on species recognition. The recognition accuracy of different classifiers was used to compare classification performance. The accuracy of the BP-RBF hybrid neural network using nine dimensional features was 96.2%, highest among all the classifiers.

KLASIFIKASI DAN IDENTIFIKASI JUMLAH KOLONI PADA CITRA BAKTERI DENGAN METODE K-NEAREST NEIGHBOR

<p class="Abstract"><em>Abstract</em> – This study proposes a system for classification and counting the number of bacterial colonies using a photo image of bacteria. The system uses several image pretreatment process. Including Contrast Stretching, <em>Extended-Maxima Transform</em>, and <em>Regionprops</em>. The main purpose of this system is to determine the category of colonies of bacteria in large quantities can not be done manually. To build the algorithms necessary features must be determined such as <em>diameter, perimeter and roundness</em> method of determining the categories using KNN <em>(K-Nearest Neighbor)</em>. As a results of this research is classify three types of bacteria such as Lactobacillus Bulgaricus, Streptococcus thermophiles, and bifidobakterium Precision with a percentage of 97,97% and 87,09% F-Measure</p><p><strong>Keywords: Contrast Stretching, Lactobacillus, Regionprops, K-Nearest Neighbor</strong></p>

Influence of Vibration on Stabilization Performance of Vehicle-Mounted Sighting Telescope

According to the effect of vibration on stabilization performance of on-board sighting telescope, a performance test analysis method is built based on the image of sighting telescope. The article mainly studies key algorithms of image pretreatment, median filtering and template matching in the process of image processing. The result indicates that vibration has significant impact on the stabilization performance of sighting telescope, especially for continuous aiming accuracy of different target. This research provides reference for the test and evaluation for the precision of vehicle-mounted sighting telescope stabilization.