Punching control process for toothed rail forge piece

Abstract

The invention relates to the field of intelligent manufacturing equipment industry, in particular to a rack rail forge piece punching control technology which comprises the steps that equipment detection data are obtained; an equipment detection coefficient is obtained according to the equipment detection data, and the equipment detection coefficient is sent to an analysis and judgment module; an equipment abnormity alarm instruction or a product hole detection instruction is generated according to the equipment detection coefficient, the equipment abnormity alarm instruction is sent to an equipment alarm module, and the product hole detection instruction is sent to a hole detection module; after a product hole detection instruction is received, the perforated toothed rail forge piece is detected, and hole detection data are obtained; a hole detection coefficient is obtained according to the hole detection data, and the hole detection coefficient is sent to an analysis and judgment module; comprehensive quality control of the toothed rail forge piece punching process is achieved, the product quality consistency is improved, and meanwhile the production cost and the quality risk are remarkably reduced.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing equipment industry, specifically to a punching control process for gear forgings. Background Technology

[0002] As a key load-bearing component in rail transit, engineering machinery, and other fields, the punching quality of gear forgings directly affects assembly accuracy, connection strength, and overall service safety. With the development of high-end equipment towards high speed and heavy load, more stringent requirements are being placed on the hole position accuracy, shape consistency, and batch stability of gear forgings. Currently, the industry generally uses automated stamping equipment for hole machining of gear forgings and supplements the production process with various inspection methods, but the following technical bottlenecks still exist: Firstly, in terms of equipment status monitoring, existing technologies largely rely on simple threshold alarm mechanisms, making it difficult to achieve early warning and accurate diagnosis of faults. Conventional methods typically determine equipment abnormalities by monitoring whether vibration amplitude exceeds a preset threshold, but this reactive alarm mode has a significant lag. Furthermore, existing monitoring systems generally lack the ability to fuse and analyze multi-dimensional vibration characteristics and predict trends, failing to identify anomalies in the early stages of faults. This leads to frequent unplanned equipment downtime and even serious accidents such as mold damage.

[0003] Secondly, in terms of hole quality inspection, traditional methods suffer from low efficiency, limited evaluation dimensions, and disconnect from equipment status. Currently, the inspection of punching quality mainly relies on manual sampling combined with two-dimensional image measurement, or the use of automated visual inspection with fixed procedures. Manual sampling is inefficient, labor-intensive, and subject to subjective errors, making it impossible to achieve 100% inspection. Summary of the Invention

[0004] In order to overcome the above-mentioned technical problems, the purpose of this invention is to provide a punching control process for gear forgings.

[0005] The objective of this invention can be achieved through the following technical solutions: A punching control process for a gear forging includes: Step 1: Inspect the laser drilling equipment used for drilling the gear forgings and obtain the equipment inspection data, which includes the vibration frequency and peak vibration acceleration of the stamping equipment within a unit stamping cycle. Step 2: Obtain the equipment detection coefficient based on the equipment detection data, and send the equipment detection coefficient to the analysis and judgment module; Step 3: Generate an equipment abnormality alarm command or a product hole detection command based on the equipment detection coefficient, and send the equipment abnormality alarm command to the equipment alarm module and the product hole detection command to the hole detection module; Step 4: Upon receiving the equipment malfunction alarm command, control the equipment to sound the malfunction alarm bell; Step 5: After receiving the product hole detection instruction, inspect the drilled gear forging to obtain hole detection data, which includes offset information and shape information, and send the hole detection data to the hole analysis module. Step 6: Obtain the hole detection coefficient based on the hole detection data, and send the hole detection coefficient to the analysis and judgment module.

[0006] As a further technical solution of the present invention: In step two, the comprehensive feature vector of the current period is input into a trained normal state single-class classifier, and the degree score of its deviation from the normal mode is output to obtain the instantaneous detection coefficient. The trend detection coefficient is obtained by calculating the linear regression slope of the sliding window mean sequence of the comprehensive feature vectors over the past 15 periods. A temporal convolutional network is trained using a temporal feature sequence of length 30 to predict the feature values ​​for the next 5 periods. The deviation between the predicted values ​​and the current normal baseline is calculated to obtain the predicted detection coefficients. The instantaneous detection coefficient, trend detection coefficient, and predictive detection coefficient are weighted and calculated to obtain the equipment detection coefficient. As a further technical solution of the present invention: In step three, if the equipment detection coefficient is greater than or equal to the equipment detection threshold, an equipment abnormality alarm command is generated and sent to the equipment alarm module. If the equipment detection coefficient is less than the equipment detection threshold, a product hole detection command is generated and sent to the hole detection module.

[0007] As a further technical solution of the present invention: In step six, the offset information and shape information are quantized, the values ​​of the offset information and shape information are extracted, and they are substituted into the weighting formula to calculate the hole detection coefficient.

[0008] As a further technical solution of the present invention: all holes on the gear forging are sequentially marked as analysis holes i, i=1, ..., n, where n is a positive integer, i is the number of any analysis hole, and n is the total number of analysis holes. The edge contour of the gear forging and the edge contour of the gear forging design drawing are obtained, and the two are superimposed to obtain the holes in the gear forging design drawing. The holes are then drawn on the edge contour of the gear forging. The position of analysis hole i and the position of the corresponding drawn hole are obtained, and the difference between the two is obtained and marked as the position offset value. The sum of the position offset values ​​of all analysis holes i is obtained and marked as offset information.

[0009] As a further technical solution of the present invention: obtain the diameter of the analysis hole i and the preset standard hole diameter, obtain the difference between the two and mark it as the diameter difference value, directly illuminate the light source above the analysis hole i, obtain the light spot area below the analysis hole i, obtain the difference between the light spot area and the preset standard light spot area and mark it as the surface difference value, sum the square of the diameter difference value and the surface difference value to obtain the hole difference value, sum the hole difference values ​​of all analysis holes i and take the average value to obtain the shape information.

[0010] As a further technical solution of the present invention: Step 7: Generate a hole abnormality alarm command based on the hole detection coefficient and obtain a defective gear forging, and send the hole abnormality alarm command to the equipment alarm module to transfer the defective gear forging to the defective area for storage, or obtain a qualified gear forging based on the hole detection coefficient and transfer the qualified gear forging to the qualified area for storage.

[0011] As a further technical solution of the present invention: Step 8: After receiving the hole abnormality alarm command, the equipment alarm module controls the hole abnormality alarm bell to sound.

[0012] As a further technical solution of the present invention: if the hole detection coefficient is greater than or equal to the hole detection threshold, a hole abnormality alarm command is generated and sent to the equipment alarm module, and the toothed rail forging corresponding to the hole detection coefficient is marked as a defective toothed rail forging, and the defective toothed rail forging is transferred to the defective area for storage.

[0013] As a further technical solution of the present invention: if the hole detection coefficient is less than the hole detection threshold, the toothed rail forging corresponding to the hole detection coefficient is marked as a qualified toothed rail forging.

[0014] The beneficial effects of this invention are: This invention achieves closed-loop monitoring of the entire process from equipment operation status to product quality output by organically combining the equipment detection module and the hole detection module. The system first performs real-time vibration analysis on the stamping equipment to provide early warning of potential faults at the equipment level; product quality inspection is only triggered when the equipment is in normal condition, avoiding invalid product inspections when the equipment is malfunctioning. This two-level linkage monitoring strategy ensures the rational allocation of inspection resources and fundamentally prevents large-scale quality problems caused by equipment malfunctions.

[0015] This invention proposes a three-dimensional detection coefficient system encompassing instantaneous, trend, and predictive aspects. By acquiring the frequency domain energy characteristics, time domain waveform characteristics, and time-series evolution characteristics of vibration signals, and employing an adaptive weighted fusion mechanism, the system can dynamically learn the normal state patterns of the equipment. This achieves a shift from reactive alarm to proactive warning. When equipment experiences progressive failures, the system can issue an early warning 45-60 seconds in advance based on the characteristic evolution trend before the vibration amplitude exceeds the limit, providing a valuable time window for preventative maintenance.

[0016] For the quality inspection of holes in gear forgings, this invention breaks through the traditional binary judgment mode of pass / fail and creatively quantifies hole quality into a calculable detection coefficient. By analyzing the positional offset and shape distortion of each hole, and considering the overall consistency of the hole group, the system can accurately identify the difference between single hole defects and systemic quality problems. Using light projection technology combined with image analysis, it can not only detect dimensional deviations but also assess deeper quality indicators such as hole wall perpendicularity and edge integrity, achieving a comprehensive and detailed evaluation of hole quality.

[0017] This invention not only provides alarm functions for abnormalities, but more importantly, it offers suggestions for problem diagnosis and resolution. When a systematic offset in the hole position is detected, the system can prompt the user to check the mold positioning device; when hole shape distortion is detected, it can suggest adjusting the stamping parameters or checking the punch wear.

[0018] In summary, the equipment status monitoring method, multi-dimensional hole quality evaluation system, and intelligent decision support mechanism of this invention enable comprehensive quality control of the punching process for gear forgings, significantly reducing production costs and quality risks while improving product quality consistency. Attached Figure Description

[0019] The invention will now be further described with reference to the accompanying drawings.

[0020] Figure 1 This is a schematic diagram of the punching control system for a toothed rail forging according to the present invention. Figure 2 This is a flowchart of a punching control method for a toothed rail forging according to the present invention. Detailed Implementation

[0021] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0022] Example 1: Please see Figure 1As shown, this embodiment is a punching control system for gear forgings, including the following modules: equipment detection module, equipment analysis module, analysis and judgment module, equipment alarm module, hole detection module, and hole analysis module; The equipment detection module is used to detect the laser drilling equipment used for drilling holes in gear forgings, acquire equipment detection data, and send the equipment detection data to the equipment analysis module. The equipment analysis module is used to obtain the equipment detection coefficient based on the equipment detection data and send the equipment detection coefficient to the analysis and judgment module. The analysis and judgment module is used to generate an equipment abnormality alarm command or a product hole detection command based on the equipment detection coefficient, and send the equipment abnormality alarm command to the equipment alarm module and the product hole detection command to the hole detection module; it is also used to generate a hole abnormality alarm command based on the hole detection coefficient and obtain a defective gear forging, and send the hole abnormality alarm command to the equipment alarm module to transfer the defective gear forging to the defective area for storage, or obtain a qualified gear forging based on the hole detection coefficient and transfer the qualified gear forging to the qualified area for storage; The equipment alarm module is used to control the equipment abnormality alarm bell to sound after receiving an equipment abnormality alarm command; it is also used to control the hole abnormality alarm bell to sound after receiving a hole abnormality alarm command. The hole detection module is used to detect the drilled toothed rail forging after receiving the hole detection command, obtain hole detection data, and send the hole detection data to the hole analysis module; wherein, the hole detection data includes offset information and shape information; The hole analysis module is used to obtain the hole detection coefficient based on the hole detection data and send the hole detection coefficient to the analysis and judgment module.

[0023] Example 2: Please see Figure 2 As shown, this embodiment illustrates a punching control process for gear forgings, comprising the following steps: Step 1: Inspect the laser drilling equipment used for drilling the gear forgings and obtain the equipment inspection data, which includes the vibration frequency and peak vibration acceleration of the stamping equipment within a unit stamping cycle. Step 2: Obtain the equipment detection coefficient based on the equipment detection data, and send the equipment detection coefficient to the analysis and judgment module; Step 3: Generate an equipment abnormality alarm command or a product hole detection command based on the equipment detection coefficient, and send the equipment abnormality alarm command to the equipment alarm module and the product hole detection command to the hole detection module; Step 4: Upon receiving the equipment malfunction alarm command, control the equipment to sound the malfunction alarm bell; Step 5: After receiving the product hole detection instruction, inspect the drilled gear forging to obtain hole detection data, which includes offset information and shape information, and send the hole detection data to the hole analysis module. Step 6: Obtain the hole detection coefficient based on the hole detection data, and send the hole detection coefficient to the analysis and judgment module; Step 7: Generate a hole abnormality alarm command based on the hole detection coefficient and obtain the unqualified gear forging. Send the hole abnormality alarm command to the equipment alarm module and transfer the unqualified gear forging to the unqualified area for storage. Alternatively, obtain qualified gear forgings based on the hole detection coefficient and transfer the qualified gear forgings to the qualified area for storage. Step 8: After receiving the hole abnormality alarm command, the equipment alarm module controls the hole abnormality alarm bell to sound.

[0024] In step two, high-frequency vibration acceleration sensors are installed at key parts of the stamping equipment. The frequency domain analysis unit uses fast Fourier transform to decompose the signal into eight key frequency bands, calculates the root mean square value of each frequency band as energy, and normalizes it to obtain the energy proportion spectrum. The peak impact peak of each stamping cycle is located by the peak detection algorithm, and the rise time slope of the peak leading edge from 10% to 90%, the fall slope of the trailing edge, and the peak half width at half maximum are calculated. The autocorrelation coefficient and moving average of the above frequency domain and time domain characteristic sequences are calculated over 20 consecutive cycles to form an evolution trajectory. The first 1000 cycles will be used as a learning period, with initial weights (e.g., frequency domain: time domain: time series = 4:3:3). After that, every 100 cycles, a lightweight gradient boosting tree model will be used to evaluate the discriminative power of each feature against recent suspected anomalies, and the weights will be dynamically adjusted. The comprehensive feature vector of the current period is input into a trained normal state single-class classifier, which outputs a score of the degree of deviation from the normal pattern, thus obtaining the instantaneous detection coefficient. The linear regression slope of the sliding window mean sequence of the comprehensive feature vector over the past 15 periods is calculated to obtain the trend detection coefficient. The larger the slope and the direction of deviation from the normal value, the higher the coefficient. A temporal convolutional network (TCN) is trained using a temporal feature sequence of length 30 to predict feature values ​​for the next 5 periods. The deviation between the predicted values ​​and the current normal baseline is calculated to obtain the predicted detection coefficients. The instantaneous detection coefficient, trend detection coefficient, and predicted detection coefficient are weighted and calculated to obtain the equipment detection coefficient; the weight factors of the instantaneous detection coefficient, trend detection coefficient, and predicted detection coefficient are 0.4, 0.35, and 0.25, respectively; the equipment detection coefficient SJ is sent to the analysis and judgment module.

[0025] In step three, the analysis and judgment module has two functions; One of its functions is to generate equipment malfunction alarm commands or product hole detection commands. The specific process is as follows: The analysis and judgment module compares the equipment detection coefficient with the preset equipment detection threshold, and the comparison results are as follows: If the equipment detection coefficient is greater than or equal to the equipment detection threshold, an equipment abnormality alarm command is generated and sent to the equipment alarm module. If the equipment detection coefficient is less than the equipment detection threshold, a product hole detection command is generated and sent to the hole detection module. The second purpose is to obtain both substandard and qualified gear forgings, and the specific process is as follows: The analysis and judgment module compares the hole detection coefficient with the preset hole detection threshold, and the comparison results are as follows: If the hole detection coefficient is greater than or equal to the hole detection threshold, a hole abnormality alarm command is generated and sent to the equipment alarm module. The toothed rail forging corresponding to the hole detection coefficient is marked as a non-conforming toothed rail forging and the non-conforming toothed rail forging is transferred to the non-conforming area for storage. If the hole detection coefficient is less than the hole detection threshold, the toothed rail forging corresponding to the hole detection coefficient is marked as a qualified toothed rail forging, and the qualified toothed rail forging is transferred to the qualified area for storage.

[0026] In step four, the equipment alarm module has two functions; One of its functions is to control the alarm bell to sound after receiving an alarm command for equipment malfunction; Its second function is to control the hole abnormality alarm bell to sound after receiving the hole abnormality alarm command.

[0027] In step five, the hole detection module is used to acquire hole detection data, which includes hole count information, offset information, and shape information. The specific process is as follows: The hole detection module sequentially marks all holes on the gear forging as analysis holes i, i=1, ..., n, where n is a positive integer, i is the number of any analysis hole, and n is the total number of analysis holes. It obtains the edge contour of the gear forging and the edge contour of the gear forging design drawing, overlaps the two, obtains the holes in the gear forging design drawing, and draws them on the edge contour of the gear forging. It obtains the position of analysis hole i and the position of the corresponding drawn hole, obtains the difference between the two, and marks it as the position offset value. It obtains the sum of the position offset values ​​of all analysis holes i and marks it as offset information. The hole detection module obtains the diameter of the analysis hole i and the preset standard hole diameter, obtains the difference between the two and marks it as the diameter difference value. A direct light source is shone above the analysis hole i to obtain the light spot area below the analysis hole i. The difference between the light spot area and the preset standard light spot area is obtained and marked as the surface difference value. The square of the diameter difference value and the surface difference value are summed to obtain the hole difference value. The hole difference values ​​of all analysis holes i are summed and the average value is calculated to obtain the shape information. The hole detection module sends offset and shape information to the hole analysis module.

[0028] In step six, the hole analysis module quantizes the offset and shape information, extracts the numerical values ​​of the offset and shape information, and substitutes them into the weighting formula to calculate the hole detection coefficient. The weighting coefficients of the offset and shape information are 0.54 and 0.46, respectively. The hole analysis module sends the hole detection coefficient KJ to the analysis and judgment module.

[0029] The working principle of this invention is as follows: This invention discloses a control method and system for drilling holes in gear forgings. The system first detects the laser drilling equipment used for drilling gear forgings using an equipment detection module to acquire equipment detection data, including vibration information. An equipment analysis module obtains an equipment detection coefficient based on the detection data. An analysis and judgment module generates an equipment abnormality alarm command or a product hole detection command based on the detection coefficient. Upon receiving the equipment abnormality alarm command, an equipment alarm module controls the alarm bell to sound. Upon receiving the product hole detection command, a hole detection module detects the drilled gear forgings using a hole detection module to acquire hole detection data, including offset and shape information. The hole analysis module obtains a hole detection coefficient based on the hole detection data. An analysis and judgment module generates a hole abnormality alarm command based on the hole detection coefficient and identifies either a defective gear forging or a qualified gear forging. Upon receiving the hole abnormality alarm command, an equipment alarm module controls the alarm bell to sound. The system first detects the laser drilling equipment to acquire equipment detection data, and then... The obtained equipment detection coefficient can comprehensively measure the degree of abnormality in the operating status of the laser drilling equipment. The larger the equipment detection coefficient, the higher the degree of abnormality. When the degree of abnormality is high, an alarm is triggered. If the degree of abnormality is low, it indicates normal operation. Then, the drilling effect is detected, and hole detection data is obtained. The hole detection coefficient obtained from the hole detection data can comprehensively measure the degree of abnormality in the quality of the drilled holes. The larger the hole detection coefficient, the higher the degree of abnormality in the hole quality. The toothed rail forgings are screened and classified according to the hole detection coefficient. This system can collect data from multiple aspects and analyze the data to detect the operating status of the laser drilling equipment and detect the quality of the holes drilled in the toothed rail forgings. It can comprehensively and accurately evaluate the quality of the holes in the toothed rail forgings, ensuring the stability of equipment operation and drilling effect. It can detect abnormalities in time and activate safety protection measures to avoid the risk of equipment damage or personnel injury, improve drilling efficiency, detect defective products in time, and adjust various parameters of the laser drilling equipment in real time to ensure drilling quality and reduce production costs.

[0030] It should be further noted that the above formulas are all derived from software simulation using a large amount of data and are selected to be close to the actual values. The coefficients in the formulas are set by those skilled in the art based on the actual situation.

[0031] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0032] The above description is merely an example and illustration of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the invention or exceed the scope defined in the claims, all of which should fall within the protection scope of the present invention.

Claims

1. A punching control process for a gear forging, characterized in that, include: Step 1: Inspect the laser drilling equipment used for drilling the gear forgings and obtain the equipment inspection data, which includes the vibration frequency and peak vibration acceleration of the stamping equipment within a unit stamping cycle. Step 2: Obtain the equipment detection coefficient based on the equipment detection data, and send the equipment detection coefficient to the analysis and judgment module; Step 3: Generate an equipment abnormality alarm command or a product hole detection command based on the equipment detection coefficient, and send the equipment abnormality alarm command to the equipment alarm module and the product hole detection command to the hole detection module; Step 4: Upon receiving the equipment malfunction alarm command, control the equipment to sound the malfunction alarm bell; Step 5: After receiving the product hole detection instruction, inspect the drilled gear forging to obtain hole detection data, which includes offset information and shape information, and send the hole detection data to the hole analysis module. Step 6: Obtain the hole detection coefficient based on the hole detection data, and send the hole detection coefficient to the analysis and judgment module.

2. The punching control process for a gear forging according to claim 1, characterized in that, In step two, the comprehensive feature vector of the current period is input into a trained normal state single-class classifier, which outputs a score of the degree of deviation from the normal pattern to obtain the instantaneous detection coefficient. The trend detection coefficient is obtained by calculating the linear regression slope of the sliding window mean sequence of the comprehensive feature vectors over the past 15 periods. A temporal convolutional network is trained using a temporal feature sequence of length 30 to predict the feature values ​​for the next 5 periods. The deviation between the predicted values ​​and the current normal baseline is calculated to obtain the predicted detection coefficients. The instantaneous detection coefficient, trend detection coefficient, and predicted detection coefficient are weighted and calculated to obtain the equipment detection coefficient.

3. The control system for drilling holes in flexible printed circuit boards according to claim 1, characterized in that, In step three, if the equipment detection coefficient is greater than or equal to the equipment detection threshold, an equipment abnormality alarm command is generated and sent to the equipment alarm module. If the equipment detection coefficient is less than the equipment detection threshold, a product hole detection command is generated and sent to the hole detection module.

4. The punching control process for a gear forging according to claim 1, characterized in that, In step six, the offset information and shape information are quantized, the numerical values ​​of the offset information and shape information are extracted, and they are substituted into the weighting formula to calculate the hole detection coefficient.

5. The punching control process for a gear forging according to claim 1, characterized in that, All holes on the gear forging are sequentially labeled as analysis holes i, i=1, ..., n, where n is a positive integer, i is the number of any analysis hole, and n is the total number of analysis holes. The edge contour of the gear forging and the edge contour of the gear forging design drawing are obtained and overlapped to obtain the holes in the gear forging design drawing. The holes are then drawn on the edge contour of the gear forging. The position of analysis hole i and the position of the corresponding drawn hole are obtained, and the difference between the two is recorded as the position offset value. The sum of the position offset values ​​of all analysis holes i is obtained and recorded as offset information.

6. The punching control process for a gear forging according to claim 1, characterized in that, Obtain the diameter of the analysis hole i and the preset standard hole diameter, and the difference between the two is marked as the diameter difference value. Directly illuminate the light source above the analysis hole i, and obtain the light spot area below the analysis hole i. Obtain the difference between the light spot area and the preset standard light spot area, and mark it as the surface difference value. Sum the square of the diameter difference value and the surface difference value to obtain the hole difference value. Sum the hole difference values ​​of all analysis holes i and take the average value to obtain the shape information.

7. The punching control process for a gear forging according to claim 1, characterized in that, Step 7: Generate a hole abnormality alarm command based on the hole detection coefficient and obtain the unqualified gear forging. Send the hole abnormality alarm command to the equipment alarm module and transfer the unqualified gear forging to the unqualified area for storage. Alternatively, obtain a qualified gear forging based on the hole detection coefficient and transfer the qualified gear forging to the qualified area for storage.

8. The punching control process for a gear forging according to claim 1, characterized in that, Step 8: After receiving the hole abnormality alarm command, the equipment alarm module controls the hole abnormality alarm bell to sound.

9. The punching control process for a gear forging according to claim 7, characterized in that, If the hole detection coefficient is greater than or equal to the hole detection threshold, a hole abnormality alarm command is generated and sent to the equipment alarm module. The toothed rail forging corresponding to the hole detection coefficient is marked as a defective toothed rail forging and transferred to the defective area for storage.

10. The punching control process for a gear forging according to claim 1, characterized in that, If the hole detection coefficient is less than the hole detection threshold, the toothed rail forging corresponding to the hole detection coefficient is marked as a qualified toothed rail forging.

Images