JPH07136109A - Endoscope apparatus - Google Patents

Abstract

PURPOSE:To achieve a speeding up of an image processing for automatic insertion or the like while reading out an image pickup signal at a high speed by connecting numerous pixels to an operation element to form a three-dimensional circuit element enabling signal processing in parallel. CONSTITUTION:A three-dimensional image sensor 22 which is integrated with an objective lens of an observation window and built into a tip part has a lamination of a light sensor section 26 as an image pickup part for photoelectric converion, a sensor switch section 27 made up of a number of sensor switch elements forming an arithmetic processing part 22B, an A/D conversion element part 28 made up of a number of A/D conversion elements and a differential circuit section 29 made up of a number of differential circuit elements. In the pixels of the light sensor section 26, for example, a common drive signal for simultaneous reading can be applied to the numerous pixels of the light sensor section 26 by controlling the sensor switch section 27 connected by a switch drive signal while analog signals subjected to photoelectric conversion can be read out simultaneously via different sensor switch elements for each of the pixels being grouped by a number of pixels.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an endoscope apparatus equipped with a three-dimensional circuit element that enables parallel signal processing by a large number of pixels and arithmetic elements.

[0002]

2. Description of the Related Art An automatic insertion system using an endoscope is
The position of the lumen is detected by dividing and determining the bright and dark parts based on the image information from the tip of the endoscope, and bending and insertion are performed toward it.

A CCD as a conventional endoscope image pickup means is
Judgment is made by sequentially reading out electrical information from the sensor line of each pixel and processing it by an image processor.

[0004]

In this case, since the speed of inserting the endoscope depends on the image processing speed, the image processing speed is too slow for the smooth insertion in the current configuration. There is.

The present invention has been made in view of the above points, and an object thereof is to provide an endoscope apparatus which can be smoothly inserted.

[0006]

SUMMARY OF THE INVENTION The present invention is an endoscope apparatus including image pickup means and calculation means for calculating image pickup information from the image pickup means, and a large number of pixels forming the image pickup means. And a large number of arithmetic elements forming the arithmetic means are electrically connected to each other to form a three-dimensional circuit element having a multi-layered structure capable of parallel signal processing, so that an imaged signal is read at high speed. In addition, the image processing for automatic insertion and the like can be performed at high speed.

That is, the conventional sequential reading method is changed to the parallel reading method, and the image processing speed is improved by the parallel image processing.

[0008]

Embodiments of the present invention will be described below with reference to the drawings. 1 to 3 relate to a first embodiment of the present invention, FIG. 1 shows an appearance of the first embodiment, FIG. 2 is a block diagram showing a configuration of the first embodiment, and FIG. 3 is a three-dimensional image sensor. The schematic configuration of is shown.

As shown in FIG. 1, the endoscope apparatus 1 of the first embodiment comprises an electric bendable endoscope 2 and an endoscope control apparatus 3 to which the endoscope 2 is connected. The endoscope control device 3 includes a calculation treatment device 3A that performs calculation of bending control, and the like.
It is composed of a monitor 3B for displaying an endoscopic image.

The endoscope 2 has an elongated insertion section 5, an operation section 6 provided at the base end of the insertion section 5, and a universal cable 7 extending from the operation section 6, and this universal unit Connect the connector 8 at the end of the cable 7 to the computing device 3
Can be connected to A. The insertion part 5 is a hard tip part 1.
1, a bending portion 12 that is formed adjacent to the tip portion 11 and is bendable, and a flexible portion 13 that has flexibility and extends from the rear end of the bending portion 12 to the tip of the operation portion 6.

For example, a joystick 14 is provided on the operating portion 6 so that the bending operation can be performed manually. By operating the joystick 14, the bending portion 12 can be bent in a desired direction. .

FIG. 2 shows a specific configuration of FIG. As shown in FIG. 2, the light guide 15 is inserted into the endoscope 2 and the connector 8 is connected to the arithmetic processing device 3A so that the illumination light of the lamp 17 in the light source unit 16 is emitted.
5 is supplied to the end face on the proximal side. The illumination light transmitted by the light guide 15 is emitted forward from the tip surface of the light guide 15 fixed to the tip portion 11 through the light distribution lens 18 attached to the illumination window to illuminate the inside of the living body.

An objective lens 21 is attached to the observation window adjacent to the illumination window, and an image of the inside of the living body is formed on the focal plane of the objective lens 21. A three-dimensional image sensor 22 as a three-dimensional circuit element is attached to the focal plane of the objective lens 21 so that the uppermost layer is exposed.

The three-dimensional image sensor 22 has a three-dimensional structure in which an image pickup section for photoelectric conversion and an arithmetic processing section 22B for driving the image pickup section and for performing signal processing on picked-up image pickup information are laminated. ing. The three-dimensional image sensor 22 is integrated with the objective lens 21 so that it can be incorporated in the tip portion 11.

The three-dimensional image sensor 22 is supplied with drive power from an image sensor drive power supply 23 via a power supply line 24. The three-dimensional image sensor 22 has a structure as shown in FIG. For example, it is a single device having a four-layer structure and made into one chip.

The three-dimensional image sensor 22 includes an optical sensor section 26 as an image pickup section for photoelectric conversion, and an arithmetic processing section 2.
A sensor switch unit 27 including a large number of sensor switch elements forming 2B, an A / D conversion element unit 28 including a large number of A / D conversion elements, and a differential circuit unit 29 including a large number of differential circuit elements. Are stacked.

The optical sensor unit 26 has a large number of optical sensor pixels arranged in a square lattice, and photoelectrically converts the optical image formed by the objective lens 21 for each pixel. Each pixel of the optical sensor section 26 is electrically connected to the sensor switch section 27 in the lower layer (schematically shown by a vertical line in FIG. 3), and the sensor switch section 27 is controlled by a switch drive signal. , For example, applying a common drive signal for simultaneous reading to a large number of pixels of the optical sensor section 26, and simultaneously reading out photoelectrically converted analog signals for a large number of pixels via different sensor switch elements for each pixel. I am able to

The sensor switch section 27 is connected to a sensor drive circuit 32 via a drive line 31 and is driven and controlled by a switch drive signal from the sensor drive circuit 32. The sensor drive circuit 32 is connected to the image sensor drive power supply 23 via a power supply line 24, and drive power is supplied from the image sensor drive power supply 23. The sensor drive circuit 32 is connected to the optical sensor unit 2 via the drive line 31.
A power supply for driving is also supplied to 6 and the like.

The A / D conversion element section 28 in the lower layer of the sensor switch section 27 is a group of A / D conversion elements for a large number of analog signals selected by the sensor switch section 27.
The data is D-converted, converted into digital data, and guided to a large number of difference circuit elements of a difference circuit unit 29 provided in the lower layer for difference processing, and a large number of digital data are simultaneously subjected to difference processing.

As described above, the three-dimensional image sensor 23
Image information is captured by the optical sensor unit 26, and the captured image information is simultaneously subjected to arithmetic processing of a large number of pixel information by a large number of arithmetic processing elements forming the arithmetic processing unit 22B (in this embodiment, arithmetic processing up to differential processing). It is supposed to be done.

The output terminal of the difference circuit section 29 is connected to an image input section 34 via a data line 33 as shown in FIG. The image data of the image input unit 34 is transferred to the video processing circuit 35 and the image calculation circuit 36. The video processing circuit 35 generates a standard video signal from the image data and displays the image captured by the optical sensor unit 26 on the monitor 3B.

On the other hand, the image calculation circuit 36 includes the optical sensor section 26.
An instruction to bend the bending portion 12 so that the image processing for calculating the center position of the dark portion in the image captured in is performed at high speed from the difference-processed image data, and the center of the dark portion is located at the center of the imaging surface. Generate a signal.

The image calculation circuit 36 is also composed of, for example, a large number of image calculation circuits, and by performing parallel image calculation processing, the center position of the dark portion can be calculated at high speed.

A feature of the three-dimensional image sensor 23 in the first embodiment is that, unlike the sequential reading method used in the conventional CCD, parallel signal reading is possible for each line of the optical sensor section 26. By doing so, high-resolution information can be output in a short time even when the number of pixels increases,
In addition, a large number of pieces of pixel information are simultaneously subjected to difference processing. Therefore, in contrast to the conventional processing in which the image processing circuit side performs processing by the sequential reading method from the CCD or serial pixel output, the desired processed output (in this embodiment) is output from the three-dimensional image sensor 23. Obtained directly. This makes it possible to perform image processing at a very high speed (for example, the processing time is about 1/10 to 1 / several hundreds) as compared with the conventional example.

The curved portion 1 adjacent to the tip portion 11 shown in FIG.
2, a bending piece (not shown) is rotatably connected to the inside of which a shape memory alloy 41 as a bending actuator is provided.
(Hereinafter, abbreviated as SMA) and angle wire 42 are built in. One end of the SMA 41 is fixed to the tip 11 via the angle wire 42,
The other end is fixed to the rear end of the curved portion.

Further, one end of each of the pair of lead wires 43 is connected to both ends of the SMA 41 so as to be electrically heated, and the other end of each lead wire 43 is energized in the arithmetic processing unit 3A. It is connected to the control circuit 44.

The energization control circuit 44 includes a pair of lead wires 4
3, the SMA driving power supply 45 and the switching element 46 are connected in series, and the switching element 46 is turned on / off by, for example, a PWM (pulse width modulation) pulse of the energization pattern generation circuit 47, and when it is on, the SMA 41 is energized. I try to heat it.

The SMA 41 is subjected to a memory process so as to be curved into a shape stored in advance by being electrically heated. Although FIG. 2 shows the SMA 41 for one-way bending, for example, the bending portion 12 is provided with similar bending actuators at four positions displaced by 90 ° along the axial direction, and the SMA 41 that is electrically heated is provided. By controlling, the distal end side of the endoscope 2 can be curved vertically and horizontally.

The input end of the energization pattern generating circuit 47 is connected to the changeover switch 48 so that the bending instruction signal selected by the changeover control section 49 is inputted.
That is, when the contact a of the changeover switch 48 is set to be turned on, the desired bending is performed by the input information from the joystick 14 as the manual bending instruction input device. Further, when the contact b of the changeover switch 48 is set to be turned on, the image calculation circuit 36 can automatically perform the bending by the input information of the automatic bending instruction.

Next, the operation of the first embodiment will be described. When the operator observes a target site such as a diseased part in the living body with the endoscope 2 or makes a diagnosis, it is necessary to insert the insertion part 5 into the living body and introduce it to a position where the target site can be observed. There is.

When performing the insertion operation by automatic insertion (automatic bending), the operator operates the changeover control section 49 to set the contact b of the changeover switch 48 to be turned on. Then, the distal end portion 11 of the insertion portion 5 is put into the oral cavity or the like, and the operation of inserting the insertion portion 5 by pushing the proximal end side or the like is performed.

In this case, the illumination light from the light source section 16 is transmitted by the light guide 15 and illuminates the inside of the living body from the tip surface thereof through the light distribution lens 18. Inside the illuminated living body, the three-dimensional image sensor 2 is formed by the objective lens 21 on the tip surface.
An image is formed on the optical sensor section 26 of the second optical sensor section 26.
Photoelectrically converted by.

The photoelectrically converted signal is the sensor drive circuit 32.
A line drive signal for driving pixels for, for example, 1 (or several) lines generated in (1) is output to the sensor switch unit 27 via the drive line 31, and the switch unit 27 outputs the line drive signal for one (or several) lines. ON / OFF of the sensor switch element is controlled so as to be commonly applied to the pixels of the minute.

Therefore, the optical sensor unit 26 is 1 (or a number).
The pixels for the lines are simultaneously driven, and the photoelectrically converted analog signal passes through the sensor switch elements for one (or several) lines of the sensor switch section 27 and then the A / D conversion element section 28.
Transferred to. Then, the pixels for one (or several) lines are simultaneously converted into digital data by the A / D conversion element section 28 and sequentially transferred to the difference circuit section 29.

In the difference circuit section 29, difference circuit elements for one (or several) lines simultaneously perform difference processing,
The difference-processed digital data is transferred to the image input unit 34 via the data line 33. The image input unit 34 transfers the transferred image data after the difference processing to the video processing circuit 35 and the image calculation unit 36, and the video processing circuit 35 displays the image captured by the optical sensor unit 26 on the monitor 3B.

Since the signal input to the video processing circuit 35 has been subjected to the difference processing, a video signal is generated by addition processing or the like. This difference processing is not particularly advantageous for image display as compared with the case where no difference processing is performed, but since the image processing does not require such high-speed processing and the data of many pixels is input in a short time. It is not an obstacle.

On the other hand, the image calculation unit 36 determines, for example, the pixel side corresponding to the dark portion of the adjacent pixels from the transferred image data after the difference processing, or compares the many difference data with the darkest portion. The pixel to which the pixel belongs is determined and the pixel corresponding to the center position is determined to be the center of the dark portion in a short time (the center position of the dark portion in the image displayed on the monitor 3B of FIG. ) Is performed), and the direction in which the dark portion is displaced from the center of the screen and the displacement amount are also calculated.

Then, an instruction signal for bending the bending portion 12 is generated so that the center of the dark portion is located at the center of the screen. In this embodiment, the SMA is set so that the dark part is located at the center of the screen.
An energization control circuit 44 sends an instruction signal for energizing and heating 41.
Output to the energization pattern generation circuit 47.

The energization heating pattern generation circuit 47 energizes and heats the SMA 41 by turning on / off the switching element 46 in order to energize and heat the SMA 41 in the corresponding direction according to the input instruction signal. Part 12
To bend. Since the curved portion 12 is curved in the direction in which the dark portion moves to the center of the screen, the tip portion 12 is always oriented in the lumen direction. Therefore, the surgeon can be introduced into the deep part of the lumen with almost no effect on the curvature of the lumen and by simply performing an inserting operation.

When the distal end side of the endoscope 2 reaches near the desired affected area, the changeover control section 49 switches the contact a of the changeover switch 48 to ON, and the joystick 14 is manually operated to make a desired bend. It becomes possible to observe the part.

From the above, since the insertion direction on the distal end side of the endoscope 2 is automatically discriminated and high-speed image processing is performed, almost real-time automatic discrimination and highly accurate insertion direction curving are performed. Can be realized, and smooth insertion can be performed compared to the conventional automatic insertion type.

As a result, the diagnosis time can be shortened, the mental and physical distress of the patient at the time of diagnosis can be significantly reduced, and the burden on the operator for the insertion operation can be reduced.

An insertion mechanism for moving the insertion portion 5 forward (for example, contacting the outer peripheral surface of the insertion portion 5 so as to sandwich it)
It is also possible to provide a pair of rollers and a motor for rotating one roller) to form an endoscope apparatus having an automatic insertion function. In this case, the rotation of the motor may be delayed when the bending instruction amount is large.

Next, a second embodiment of the present invention will be described. Second
The configuration of the embodiment is almost the same as that of the first embodiment, and the different point is the configuration of the three-dimensional image sensor 51. FIG. 4 shows a three-dimensional circuit element used in the second embodiment.
The structure of the three-dimensional image sensor 51 is shown.

The uppermost layer is the photoelectric conversion element section 52, from which image input information is fetched. In the next second layer, the level detection circuit section 53 is arranged, and the photoelectric conversion element section 5 is provided.
The digital signal is obtained by converting the analog electric signal obtained in step 2.

The third layer is a first frame memory 54 for storing each detection signal, and the fourth layer is a second frame memory &
The arithmetic circuit unit 55. Further, a control circuit unit 56 for controlling transfer and processing of signals between the respective layers is connected to the respective layers. The control circuit unit 56 is composed of a subtraction circuit and the like. This three-dimensional image sensor 51
Is also a device in which a four-layer structure is made into one chip as in the first embodiment.

With the above configuration, for example, image data when the endoscope 2 is inserted is stored in advance in the second frame memory of the second frame memory & arithmetic circuit unit 55. Next, at the next image input time, new image data is input to the first frame memory 54. Therefore, the image data stored in the second frame memory and the first captured image data
Data is subtracted from the frame memory 54.

This makes it possible to detect how much the positional deviation has occurred from the previous information. After performing the above processing, the image input data is stored in the second frame memory. By performing the above processing for each image input, the insertion direction of the endoscope 2 can be detected in real time.

At the time of the subtraction process, the direction of insertion in the lumen can be found by tracing the dark part as in the first embodiment.

With the above-described structure, as in the first embodiment, the position shift is curved in the direction of the positional deviation. This facilitates automatic insertion into the lumen. The three-dimensional image sensor 51 described above is also the first
Since parallel processing of read data is performed as in the embodiment, high-speed image processing can be performed. This enables smooth automatic insertion, shortens the diagnosis time, and reduces the pain to the patient.

Next, a third embodiment of the present invention will be described. In the intraluminal diagnosis, if automatic insertion is possible, it is possible to remember the image data of the site once the examination is performed and automatically guide the endoscope to the site. Conventionally, in endoscopic diagnosis, the position of the patient is recognized by X-rays for each examination, but if this can be eliminated, the exposure of the patient and the operator can be reduced. In this embodiment, it is possible to reduce the X-ray exposure.

The constitution of this embodiment is almost the same as that of the second embodiment, but is characterized in that a memory for comparing patient image information stored in advance in the three-dimensional image sensor 60 is added. ing. The configuration of the three-dimensional image sensor 60 used in this embodiment is shown in FIG.

The photoelectric conversion element section 61 is provided in the uppermost layer, and image input information is taken in from this. In the next second layer, a level detection circuit section 62 is arranged, which converts the analog electric signal obtained by the photoelectric conversion element section 61 into a digital signal. The third layer is a first frame memory 63 that stores each detection signal, and the fourth layer is a frame memory & arithmetic circuit unit 64.

As a fifth layer, a third frame memory 65 for storing image data of a desired affected area is arranged. Here, the image data stored in the third frame memory 65 needs to be inserted into the affected area once and the image data must be saved. A control circuit unit 66 for controlling transfer and processing of signals between the layers is connected to each layer. The control circuit unit 66 is composed of a subtraction circuit and the like.

A method for performing automatic insertion in the above configuration will be described. The image processing method for each image capture is the same as in the second embodiment, but the capture information (first
The data stored in the frame memory 63) and the image data of the affected area previously stored in the third frame memory 65 are also compared. If the same image information is recognized in this comparison, it means that the endoscope can approach the desired site.

From the above, once the image information from the endoscope 2 is stored, the affected area can be automatically approached from the second examination, so that the affected area can be checked by X-ray every time. You no longer need what you were doing. This makes it possible to reduce the X-ray exposure of the patient and the operator.

In the first embodiment, the image pickup means is attached to the tip portion 1.
The case of the electronic endoscope housed in the first embodiment has been described, but the present invention is not limited to this, and the three-dimensional image sensor 22 is attached to the TV camera detachably attached to the eyepiece of the fiberscope. It can also be applied when stored. Also, 3
The three-dimensional image sensor 22 is not limited to the configuration shown in FIG.

In the above-described embodiment, an example in which a signal for determining the bending amount of the bending portion for facilitating insertion of the insertion portion is generated from the imaged signal in a short time has been described. For example, the present invention can also be applied to the case where image processing for generating a three-dimensional image from captured two-dimensional image information is performed in a short time, or when other operations or functions are controlled.

[0059]

As described above, according to the present invention, since high-speed arithmetic processing can be performed, automatic insertion of an endoscope or the like can be performed smoothly and in a short time. The diagnosis time can be shortened. In addition, it is possible to significantly reduce the mental and physical pain at the time of diagnosis for the patient.

[Brief description of drawings]

FIG. 1 is an overall configuration diagram showing an appearance of a first embodiment of the present invention.

FIG. 2 is a block diagram showing the configuration of the first embodiment.

FIG. 3 is an exploded perspective view showing a schematic configuration of a three-dimensional image sensor.

FIG. 4 is an exploded perspective view showing a schematic configuration of a three-dimensional image sensor according to a second embodiment of the present invention.

FIG. 5 is an exploded perspective view showing a schematic configuration of a three-dimensional image sensor according to a third embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Endoscope device 2 ... Endoscope 3 ... Endoscope control device 3A ... Arithmetic processing device 3B ... Monitor 5 ... Insertion part 12 ... Bending part 14 ... Joystick 15 ... Light guide 21 ... Objective lens 22 ... 3D image Sensor 22B ... Calculation unit 26 ... Optical sensor unit 27 ... Sensor switch unit 28 ... A / D conversion element unit 29 ... Difference circuit unit 31 ... Drive line 32 ... Sensor drive circuit 33 ... Data line 35 ... Video processing circuit 36 ... Image calculation Circuit 41 ... SMA 42 ... Angle wire 44 ... Energization control circuit 45 ... Energization heating power supply 46 ... Switching element 47 ... Energization pattern generation circuit 48 ... Changeover switch 49 ... Changeover control unit

Claims (1)

[Claims] 1. An endoscope apparatus comprising an image pickup means and a calculation means for calculating image pickup information from the image pickup means, wherein a large number of pixels forming the image pickup means and a large number of pixels forming the calculation means are provided. An endoscope apparatus characterized in that it is formed of a three-dimensional circuit element having a multi-layered structure which is electrically connected to an arithmetic element and can perform signal processing in parallel.