RESUMO
Flocks of birds may cause major damage to fruit crops in the ripening phase. This problem is addressed by various methods for bird scaring; in many cases, however, the birds become accustomed to the distraction, and the applied scaring procedure loses its purpose. To help eliminate the difficulty, we present a system to detect flocks and to trigger an actuator that will scare the objects only when a flock passes through the monitored space. The actual detection is performed with artificial intelligence utilizing a convolutional neural network. Before teaching the network, we employed videocameras and a differential algorithm to detect all items moving in the vineyard. Such objects revealed in the images were labeled and then used in training, testing, and validating the network. The assessment of the detection algorithm required evaluating the parameters precision, recall, and F1 score. In terms of function, the algorithm is implemented in a module consisting of a microcomputer and a connected videocamera. When a flock is detected, the microcontroller will generate a signal to be wirelessly transmitted to the module, whose task is to trigger the scaring actuator.
Assuntos
Inteligência Artificial , Frutas , Animais , Aves , Redes Neurais de ComputaçãoRESUMO
Removing noise from an free induction decay (FID) signal is of fundamental significance in the analysis of results of magnetic resonance (MR) spectroscopy and tomography. Optimum solution can be seen in removing noise by means of a digital filter bank that uses half-band mirror frequency filters of the type of low-pass and high-pass filters. A filtering method using digital filters and the approach threshold adjustment is described in the paper.
RESUMO
Imaging techniques based on the principles of nuclear magnetic resonance (NMR) are modern techniques for the study of chemical, biological and physical properties of substances. The most important are their applications in medical sciences. MR imaging of a specimen weighted with diffusion coefficients requires very accurate data on the time course of the gradient pulse. Diffusion coefficients are determined from the drop of the MR signal measured with and without the application of magnetic field gradients. From the accuracy point of view, the defined course of gradients plays an important role in the computation of coefficients. A minimum rise and fall times, a defined magnitude of the excited gradient of the magnetic field and a symmetry of positive and negative pulses (zero integral of pulses of the same magnitude and opposite polarities) are required. To characterize the time course of gradient pulses or either polarity, simple methods of their measurement has been developed and experimentally tested on a 4.7 T tomograph. The method is based on the principle of instantaneous MR frequency measurements at the presence of a gradient pulse following the excitation of a thin layer situated outside the centre of the gradient field.