Study On The Number Of Elevator Car Based On Digital Image Process

Energy conservation and emissions reduction is the countries vigorously advocation, building energy conservation is imperative, mainly include two aspects to solve, one is "passive energy saving"; the second is "actively energy saving", the latter according to the personnel presence is the way to effectively achieve overall energy saving in building. Using the infrared device can counts the numbers by the stairs, but can not statistical the elevators, therefore this paper puts forward a kind of counting method based on image processing of the elevator car. It provides an effective method to solve the numbers of people in the elevator car. Video images of elevator are derived with CCD camera and the image pretreatment, head feature extraction, detection, counts can be getting through the digital image processing. It can be further optimized the electricity utilization energy-saving control strategies between the construction equipment and users. In order to count up the number of people, the first is the characteristic of head extracting, therefore, the content of the paper research mainly includes two sides:According to the features of targets in the elevators, the image pretreatment is researched. This paper gives a kind of moving target detection algorithm based on background modeling background subtraction on the basic of comparative the advantages and disadvantages of algorithm, Adopting adaptive threshold binary image, image segmentation and Canny operator edge detection and median filter for denoising to get the complete head characteristics of the binary image, head feature extraction to lay the foundation for the next step.The preparation of head detection and counting is extracting the features from binary image, By studying the shape、color、texture feature extraction algorithm, the paper using chain code algorithm to extract the circle outline based on hair color, shape features. On the basis of existing personnel detection algorithm, proposed the algorithm based on Hough transform detection circular object and morphological markers connected component. The algorithm can detects the image of circular head contour and mark the upper part of the human body contour connected component. The experiments prove the algorithm can counts the number of people effectively.The method of counting the number of elevator bases on the digital image processing provides a way to solve the number of people statistics in the high-building. This method remedies the defect of the infrared sensor to improve the detection accuracy and optimize the control strategy has practical research significance.