A Convolutional Neural-Network-Based Training Model to Estimate Actual Distance of Persons in Continuous Images.

Distance and depth detection plays a crucial role in intelligent robotics. It enables drones to understand their working environment to avoid collisions and accidents immediately and is very important in various AI applications. Image-based distance detection usually relies on the correctness of geometric information. However, the geometric features will be lost when the object is rotated or the camera lens image is distorted. This study proposes a training model based on a convolutional neural network, which uses a single-lens camera to estimate humans' distance in continuous images. We can partially restore depth information loss using built-in camera parameters that do not require additional correction. The normalized skeleton feature unit vector has the same characteristics as time series data and can be classified very well using a 1D convolutional neural network. According to our results, the accuracy for the occluded leg image is over 90% at 2 to 3 m, 80% to 90% at 4 m, and 70% at 5 to 6 m.

Assessment of treatment alternatives for laboratory cod wastewater: a practical comparison with emphasis on cost and performance.

This study focused on investigation of treatment alternatives for COD wastewater from academic laboratories, using a number of technologies including chemical reduction/precipitation, ion exchange and adsorption by chitosan. Results showed that high concentrations of 375 mg l(-1) chromium, 1,740 mg l(-1) mercury and 993 mg l(-1) silver in COD wastewater can be reduced to 2.34 mg l(-1), 3.65 mg l(-1) and 1.89 mg l(-1) respectively, by the chemical reduction/precipitation process. Results from ion exchange at a flowrate of 20 ml min(-1) showed breakthrough effluent concentrations obtained at 0.59 mg l(-1) chromium, 3.92 microg l(-1) mercury and 0.65 mg l(-1) silver corresponding to 75.6 l at 63 hr, 40.8 l at 34 hr and 33.6 l at 28 hr respectively. Kinetic and isotherm studies revealed that chitosan can adsorb Cr6+, Hg2+ and Ag+ ions most effectively at a flowrate of 20 ml min(-1) and the optimum pH for feed solution is 4. Chitosan column experiments indicated that average effluent concentrations at breakthrough point for chromium, mercury and silver are 0.76 mg l(-1), 6.04 mg l(-1) and 0.51 mg l(-1) respectively with throughput volumes and retention times of 120 l at 100 hr, 60 l at 50 hr and 48 l at 40 hr. Results of solidification experiments for chemical sludge and residual chitosan based on compressive strength and metal leachability tests showed, that the acceptable ranges of the solidification parameters were: sludge/cement = 0.1-1.0 (weight/weight), water/cement = 0.5-0.6 (weight/weight) and sand/cement = 0.5-3.0 (weight/weight). Operating cost per litre of COD wastewater treated, based on the current prices in Thailand was found to be Baht 19.95 for the chemical reduction/precipitation process, Baht 96.35 for ion exchange treatment and Baht 18.29 for chitosan adsorption.