The invention belongs to the technical field of machine vision detection, and discloses a machine vision based device and method for detecting a flotation liquid level on line, which are particularly applicable to technological processes of light floatation and heavy floatation of potassium liquid. The device comprises an observation window (1), an industrial camera (2), an Ethernet network (3), an industrial personal computer (4), a complementation light source (5), a programmable logic controller (6), a scraper mechanism (7), a floating overflow groove (8), a cleaning device (9) and a floating solution (10). The industrial camera collects images of floating liquid and foams from the observation window and then sends the images to the industrial personal computer, and the industrial personal computer performs liquid level extraction according to collected image information with a machine vision based analysis method. The device and the method can perform efficient liquid level extraction on the floating liquid and the foams and have the advantages of accurate detection, stable operation, lower power consumption of a scraper and the like.

Machine vision based device and method for detecting flotation liquid level on line

Sulfur flotation liquid level measuring method based on foam image characteristic and air volume

Self-adaptation flotation froth image enhancement method based on fractional order differential

The invention relates to a coal flotation process automatic control system and method. The coal flotation process automatic control system comprises a flotation machine, an ore slurry pre-treatment machine, a colloidal mill, a collecting agent storage tank, a frothing agent storage tank, detection equipment and executing equipment, wherein the colloidal mill is connected with each of the flotation machine and the ore slurry pre-treatment machine; the frothing agent storage tank is connected with each of the flotation machine and the ore slurry pre-treatment machine; the detection equipment connected with the executing equipment is arranged in the flotation machine. According to the coal flotation process automatic control system and method, a video camera is utilized to perform real-time detection on flotation froth and floatation tailings subjected to pressure filtration, so that the image information of froth and tailings subjected to pressure filtration can be collected and uploaded to an industrial personal computer, and meanwhile real-time state feature parameter values of the froth and tailings images are extracted and compared with a given threshold value determined by an off-line standard experiment, so as to obtain the froth information of enrichment clean coal and the coal leakage quantity information of tailings; besides, the dosage, the clear water quantity and the air input are adjusted in real time through the industrial personal computer according to the froth information and the coal leakage quantity by adopting the control method, the flotation machine can maintain a better working state, and a stable product quality can be ensured.

Coal flotation process automatic control system and method

A zinc flotation concentrate grade soft measurement method based on a procedural texture characteristic is provided. The method combines expert knowledge and a data modeling process. That a foam imageis represented by a single-frame texture characteristic based on image statistical characteristics is put forward according to key points of foam observation by on-site workers; digitalization of thecurrent production state with a texture sequence is put forward according to a characteristic that the current production state is determined by observation of the foam state in a period of time by on-site workers; and a modeling method for the texture sequence is provided to reduce the dimensionality of a feature vector. An improved boosting decision tree algorithm is adopted in a prediction algorithm to effectively inhibit an overfitting problem caused by too-fast learning, and improving generalization capability. Experiments prove that the method is simple in calculation, high in implementation speed and high in prediction accuracy, actual operation on site is convenient so as to direct spot operation in real time to optimize a production process, and a problem that on-line detection of the grade of zinc ore is difficult is solved.

A zinc ore grade soft measurement method based on a procedural texture characteristic

A visual monitoring device for a flotation froth image is used in the beneficiation process monitoring and the optimization control field. The monitoring device consists of a high-resolution industrial camera, a high-frequency light resource and hardware equipment like a computer. By debugging and installing various hardware equipment, particularly installing a double protective cover, the device reaches IP65 protection class and meets the requirements of acquiring a high-quality froth image and real-time processing in the respect of the device performances. The image acquisition method is as follows: the industrial camera acquires an image of a flotation tank froth layer; video signals are digitized, and digital video signals are generated and transmitted to an image conversion card by an optical fiber interface and an optical fiber. The image conversion card converts the protocol data of the camera into the digital signals which are suitable for computer processing and reads the data to a memory of the computer by a hardware driver interface, and an image processing and application program reads, analyzes and processes the froth image.

Xiaojun Shen

Papers

Floatation foam image vision monitoring apparatus