JPS6315377A - High speed image input device - Google Patents

Abstract

PURPOSE:To eliminate a loss of a processing time by stopping a sample moving base and inputting an image, by providing an object tracking means for tracking an object by a sample moving means and forming an object image by a high resolution image input means. CONSTITUTION:As for a method for constituting an object follow-up means, for instance, one movable mirror, and the other movable mirror are allowed to have the degree of freedom corresponding to the vertical direction of the moving direction 104 of an object (sample moving base), and the degree of freedom corresponding to the parallel direction, respectively, the former movable mirror is moved by an angle portion corresponding to a displacement 106 of an object 100 which has inputted an image immediately before by a high resolution image input means and a detected object 100', and the latter movable mirror is moved at an angular velocity corresponding to a moving speed of the object 100', by which an image of the object can always be formed as an image which has stood still, in the high resolution image input means. In this way, a high resolution image input is executed without stopping the sample moving base, and an image input processing can be executed at a high speed.

Description

[Detailed Description of the Invention] [Industrial Application Field] The invention relates to image input of objects sparsely distributed in a specimen, and particularly to an image input device suitable for high-speed input of white blood cell images in a blood specimen. .

[Conventional technology]

Conventional devices use a one-dimensional line sensor (CCD array, etc.) as a low-resolution image input means, and a two-dimensional surface sensor (image pickup tube, image pickup plate, etc.) as a high-resolution image input means. Move the specimen table with the specimen in the vertical direction.

When the object crosses the one-dimensional line sensor, the object is detected, the object is moved to the center of the input area of the high-resolution image input means in a known positional relationship, and the specimen moving stage is stopped. input the image of the object. This did not take into account processing time losses caused by sequentially performing low-resolution painter input and high-resolution image input, and by stopping the sample moving table one by one.

Furthermore, in Japanese Patent Application Laid-Open No. 51-77391, ``Method and Apparatus for Analyzing Background with Two Levels of Resolution,'' two surface sensors with a known positional relationship are used as a low-resolution image input means and a high-resolution image input means, respectively. Although this method aims to speed up processing by parallelizing image input at two resolutions, no consideration was given to the loss of processing time caused by stopping the specimen moving stage one by one.

[Problem that the invention seeks to solve]

The above-mentioned conventional technology does not take into consideration the loss of processing time due to the stoppage of the specimen moving table, and the speed of image input of the target object is not always sufficiently high.

An object of the present invention is to perform high-resolution image input without stopping the specimen moving stage and to speed up image input processing.

[Means for resolving the intermediate point] The above purpose is to use a one-dimensional sensor whose image input area is the area 102 as a low-resolution image input means, as shown in the visual field diagram 101 in FIG. This is achieved by moving the specimen moving table in the vertical direction and providing an object tracking means that tracks the movement g:h104 of the object 100 and moves the image/image input area 103 of the high-resolution image input means. In other words, when viewed from the high-resolution image input means, the object 100 is
This is achieved by creating a state of relative stillness.

[Effect]

The object tracking means can be constructed by combining two movable mirrors in front of the high-resolution image input means and swinging one part of the object two-dimensionally. There is a method of shifting the optical axis by moving a one-quadrant lens two-dimensionally on a plane perpendicular to the optical axis.

For example, in the former method, the object (
The free length corresponding to the vertical direction of the moving direction 104 of the sample moving table (specimen moving table) is given a free length corresponding to the parallel direction to the other movable roller, and the object 100 whose image was inputted immediately before by the high-resolution image input means is By moving the former movable mirror by an angle corresponding to the displacement 106 of the detected object 100' and moving the latter movable mirror at an angular velocity corresponding to the moving speed of the object ioo', the image of the object is Always high resolution -1
The image can be formed as a stationary image on the image input means.

〔Example〕

An embodiment of the present invention will be described below with reference to FIG.

This apparatus includes a specimen moving table 10 that carries a specimen and moves it, an objective lens 21 that magnifies the image of the object. A semi-transparent mirror 30 that splits the light into a low-resolution painter input means and a high-resolution image input means. Low magnification lens 22 for inputting low resolution images
．． An optical filter 41, a -dimensional line sensor 51, an object detection circuit 60, a high-magnification lens 23 for inputting high-resolution images, an optical filter (or color separation system) 42. two-dimensional surface sensor) 53
．． A/D circuit 65 A control circuit 7 for tracking the object and forming a stationary object image on the high-resolution image input means
0, Drive circuit 81, 82° 83, Movable mirror system 90, 9
1.92.

A drive circuit 81 is operated in response to a signal from the control circuit 70, and the sample moving stage 10 is moved at a constant speed in a direction 104 corresponding to the vertical direction of the -dimensional line sensor.

The specimen image on the specimen moving stage 10 is magnified by the objective lens 21,
The enlarged image is divided by a semi-transparent mirror 30 and one part is guided to the low magnification lens 22 side. On the other hand, the light that passes through the low magnification lens 22 is guided to a movable mirror system 90, passes through an optical filter 41 that emphasizes the contrast between the object and the background, and is then guided to a -dimensional line sensor 51 where it is converted into an electrical signal. Ru. The output signal of the above-mentioned -dimensional line sensor 51 is guided to an object detection circuit 60, and the object detection circuit 60 detects the object based on a preset darkness value or a threshold value calculated from the above output signal. Inspect for presence.

If no target object is detected by the target object detection circuit 60, only the sample moving table 10 continues to move. If an object is detected, the one-dimensional line center 51 of the object is detected.
The above position information is transmitted from the object detection circuit 60 to the control circuit 7.
0, the control circuit 70 calculates the displacement between the position of the object whose image was input just before and the above position information, calculates the rotation angle corresponding to the displacement, and operates the drive circuit 82 to move the movable mirror. 91 to a predetermined position. Simultaneously with the above control, the displacement of the object from the image input area 102 of the one-dimensional line sensor and the angular velocity corresponding to the moving speed of the specimen moving table 10 are calculated, the driving rotation 83 is operated, and the movable mirror 92 is moved to a predetermined position. After rotating to this position, the rotation continues at the above angular velocity.

After the image of the object captured by the movable mirror system 90 (movable mirrors 91 and 92) is highly magnified and enlarged by the zoom lens 23, it is passed through an optical filter (or color separation system) 42 and received by the three-dimensional surface sensor 52. The output signal of the two-dimensional surface sensor 52 is guided to a sample/hold and A/D conversion circuit 65,
Image input is executed by a signal from the control circuit 70.

When multiple objects are detected consecutively or simultaneously, information regarding the (relative) position of the object is held in the control circuit 70 for objects for which high-resolution images have not yet been input, and images are sequentially input. Ru.

The above is the operation of the embodiment. Here, the control circuit 70 can be realized by a general-purpose CPU, and the two-dimensional surface sensor 52 may be one or more image pickup tubes, one or more image pickup plates, or the like.

Also, as a modification of the embodiment.

1. The order of the movable mirror system 90 and the high magnification lens 23 is changed. 2. An electronic shutter that can turn the transmission of light on and off is provided between the semi-transparent mirror 30 and the two-dimensional surface center 52 to move the specimen. Stand 10. 3. A configuration in which the influence of the vibrations of the movable mirror system 90 is reduced. 3. A configuration in which the influence of the vibrations is reduced by using an element that can be forcedly read out at any time as the three-dimensional surface sensor 52. 4. Appearance of the object. Examples include a configuration in which the control circuit 70 has a function of adjusting the speed of the specimen moving table 10 and the movable mirror 92 according to the interval, and controlling the image input at a speed that matches the density variation of the target object distribution within the specimen. It will be done.

〔Effect of the invention〕

According to the present invention, the image of the object can be formed stationary on the high-resolution image input means without stopping the object (specimen moving stage), so the image can be inputted while stopping the specimen moving stage. It is possible to reduce the loss of processing time due to this, and has the effect of speeding up the input of images of objects.

[Brief Description of the Drawings] Fig. 1 shows an embodiment of the invention, and Fig. 2 shows the i-image input area of the high-resolution image input means and the low-resolution image input means on the field of view of the optical system. It is a diagram showing a relative relationship. 10... Specimen moving table, 21, 22.23... Lens. 30... Semi-transparent mirror, 41... Optical filter, 4
2... Optical filter (or color separation system), 51...
・-dimensional line sensor, 52... two-dimensional surface sensor,
60...Target detection circuit, 65...A/D circuit, 7
0...Control circuit, 81,82.83...Drive circuit,
90... Movable mirror system, 91.92... Movable mirror, 100° 100'... Target image, 101... Optical field of view, 102... Image input area of low resolution image input means, 103. ... Image of high-resolution image input means (elephant input area, 104... Movement direction of object (specimen moving stage), 1
05...Area where image input is possible, 106...Displacement between objects. Agent Patent attorney Katsuo Ogawa, -+7:',:'l (゛ ・7 ゝ, -1

Claims (1)

[Claims] 1. In a test specimen in which the objects are sparsely distributed, the present invention has a low-resolution image input means for detecting the object and a high-resolution image input means for inputting the object image. a specimen moving means for moving the object relative to the low-resolution image input means; and object tracking for tracking the object moved by the specimen moving means and forming a target image on the high-resolution image input means. 1. A high-speed image input device, comprising: means. 2. The high-speed image input device according to item 1, which includes instantaneous light transmission means for instantaneously connecting and disconnecting the optical path as part of the high-resolution image input means. 3. The first method is characterized in that the speed of the specimen moving means and the object tracking means is changed by the output of the low resolution input means.
The high-speed image input device described in Section 1. 4. The high-speed image input device as set forth in item 1, characterized in that it simultaneously comprises the means described in items 2 and 3.