The present innovation relates to a method of determining particle size information of particles, in particular moving particles, such as pharmaceutical particles.
U.S. Pat. No. 7,733,485 discloses a method for measuring the size and shape of powdery and grain like particles. The method requires sample preparation for leveling the measured particle surface and taking at least two pictures, which excludes the method from measuring moving particles. In the method at least two images of the surface of the sample are taken under illuminations from two light sources placed symmetrically on each side of the sample, the surface structure of the sample remaining at least essentially unchanged in the different measurements. Thus, the method is not suitable for measuring moving particles. In the method described in U.S. Pat. No. 7,733,485 the distance of the measurement head to the particles is made constant by using a transparent glass plate on which the particles are supported.
Other disclosed methods for particle size measurement, such as that disclosed in US 20040151360, typically need special sample preparation. This clearly excludes them from measuring moving particles. Other methods use indirect measurement, where a feature is measured which is only related to particle size. Such methods are disclosed in U.S. Pat. No. 5,870,190 and U.S. Pat. No. 7,420,360. This requires learning of the relation, which can be laborious and valid only for particles and materials that are used for creating the relation.
It is an aim of the invention to provide a particle size measurement method and system which is suitable not only for stationary but also for moving particles and does not involve any sample preparation.
A further aim is to accomplish a method and system for particle size measurement without the need to have a transparent glass plate on which the particles are supported. The invention is based on the idea of capturing images of the particles by combining three or more different illumination geometries to the color channels of the image. Further, the images are processed.
More specifically, the present method and system are defined in the independent claims.
According to one aspect, the method for determining particle size information of particles contained in a sample comprises:
According to one embodiment, the output of the detector forms an image in which some of the particles are in focus and some are out-of-focus, because of suitable detection optics. The particles that are in focus are detected by image processing.
According to one embodiment, the light sources are placed around the sample at equal angles α, as in
According to one embodiment, the sample is a fluid, preferably a powder, liquid or gas, which contains the particles to be measured. According to one embodiment, the sample is a non-fluid powder surface containing particles. The particles may be pharmaceutical particles. The fluid may be in constant motion. Thus, it may form a stream, which allows for continuous on-line monitoring of processes.
According to one embodiment, there is provided a transparent window between the detector/light sources and the sample. The transparent window may comprise a glass plate. According to one embodiment, there is provided at least two superimposed glass plates, by which arrangement a wider angle of illumination can be achieved should the window on the sample side be small or have a light-blocking collar.
According to one embodiment, the focal plane of the camera is at the surface of the transparent window. This is a preferred option at least when the sample is opaque, such as powders typically are. Thus, the measurement is directed to the topmost layer or layers of the sample.
According to another embodiment, the focal plane of the camera is in the sample at a distance from the transparent window. The depth of field of the image obtained and subsequent image processing can used to take pick the particles which are in focus for particle size determination. Said processing may comprise
Thus, particles not in focus based on predefined criteria are preferably not used in the size determination. The depth of field of the imaging optics forms a constant volume. This information can be used for determining the density of the particles of interest in the sample.
Determination of the particles that are in focus helps to avoid size determination problems associated with at least some known optical size measurement methods.
According to one embodiment, the illumination is realized as a short and high energy light pulse, for eliminating particle movement. In this case, the detector can be a camera whose exposure time is adjusted to be longer than the illumination period of the light sources.
This helps to perform the measurement for fast-moving particles as the required temporal capabilities of light sources are generally better than those of detectors. Preferably, the light sources are capable of providing a light pulse having a duration of less than 5 μs.
The term particle is herein used to describe solid particles, droplets, bubbles or other visually identifiable entities. The sample may be essentially formed by the particles themselves or it may contain a fluid or solid medium as a carrier of the particles.
Considerable advantages are obtained by means of the invention and its embodiments. First, no sample preparation, such as leveling, is needed. Second, there is no need to have a constant distance between the measurement head and the particles. Third, the size estimation is based on estimating the actual geometrical size of the particles rather than indirect estimation like previous methods. The movement of the particles is compensated with a very short and high energy light pulse, which enables capturing images without motion blur. Fourth, the disclosed method also allows one to measure through a small diameter glass plate where the boundary would block the illumination with traditional geometry. Moreover, the illumination geometry can be such that particles can be measured even behind a thick glass.
Next, embodiments of the invention are described in more detail with references to the attached drawings.
The present invention can be used for imaging particle-containing fluids, including fluid powders, or non-fluid particulate matter. In addition to powdery form, the fluid may be in gaseous or liquid form. The fluid is typically provided in the form of a stream having a velocity. Thus, the particles contained in the fluid may be moving during the measurement.
The camera can be, for example, a CCD camera or other digital imaging unit capable of both spatial and spectral resolution.
The illumination units may comprise any units known per se. Preferably, they are capable of providing pulsed light, which may be achieved using an inherent property of the light source (flash-type lamps) or a separate light chopper (continuous-type lamps). Similarly, the wavelength channel of each of the light sources can be inherent to the light source itself (e.g. LED lamps or other narrow-band light sources) or achieved using filters provided in the light paths (broadband light sources). The three or more light sources may also distribute light originating from a single parent light source, for example, using optical waveguides, such as optical fibers, or reflecting elements, such as prisms or mirrors.
The wavelength bands of the light sources naturally should overlap, at least partly, with the detection channels of the detector. According to one embodiment, the wavelength bands and detection channels correspond to red, green and blue wavelengths.
In addition to the parts listed above, the system preferably comprises a control unit (not shown) for automatic control of the exposure time of the camera and for illumination of the sample by the illumination units. In addition, there may be analysis means (not shown), such as a computer, for image capture, optionally also image storage, and signal processing.
The present method does not necessitate the presence of a glass plate or any other transparent window. It can be applied also directly for free, stacked, floating, dropping or sprayed particles, for example, without a window separating the illumination units/imaging optics and the sample. The imaging optics define the zone of interest, which is at a constant distance from the detector with predefined optical arrangement.
Using different wavelengths for different illumination angles enables capturing the different illumination geometries with a single color image. Using this, the color values encode the surface gradient information, which be transformed into 3-D information following the well known principle of photometric stereo (“Photometric Method for Determining Surface Orientation from Multiple Images.” Woodman, Robert J. Optical Engineering, Vol. 19, No. 1, 1980, pp. 139-144). The present invention takes advantage of the 3-D information in the particle detection stage of the image processing and particle size estimation. Using only gray level intensity information, occluded particles can lead to false detections, where e.g. a group of particles is determined as one, if there is not enough shadow or gradient to provide information for separating particles form each other. E.g. in such cases the 3-D information gives additional information and leads to more precise detections.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/FI2012/050029 | 1/13/2012 | WO | 00 | 9/9/2013 |
Number | Date | Country | |
---|---|---|---|
61433982 | Jan 2011 | US |